Large dynamic factor models, forecasting, and nowcasting in Statsmodels

statespace_large_dynamic_factor_models.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Large dynamic factor models, forecasting, and nowcasting\n",
    "\n",
    "Dynamic factor models postulate that a small number of unobserved \"factors\" can be used to explain a substantial portion of the variation and dynamics in a larger number of observed variables. A \"large\" model typically incorporates hundreds of observed variables, and estimating of the dynamic factors can act as a dimension-reduction technique. In addition to producing estimates of the unobserved factors, dynamic factor models have many uses in forecasting and macroeconomic monitoring. One popular application for these models is \"nowcasting\", in which higher-frequency data is used to produce \"nowcasts\" of series that are only published at a lower frequency.\n",
    "\n",
    "**Table of Contents**\n",
    "\n",
    "This notebook describes working with these models in Statsmodels:\n",
    "\n",
    "1. [Brief overview](#Brief-overview)\n",
    "2. [Dataset](#Dataset)\n",
    "3. [Specifying a mixed-frequency dynamic factor model with several blocks of factors](#Model-specification)\n",
    "4. [Model fitting / parameter estimation](#Model-fitting-/-parameter-estimation)\n",
    "5. [Estimated factors](#Estimated-factors)\n",
    "6. [Forecasting observed variables](#Forecasting)\n",
    "7. [Nowcasting GDP, real-time forecast updates, and a decomposition of the \"news\" from updated datasets](#Nowcasting-GDP,-real-time-forecast-updates,-and-the-news)\n",
    "8. [References](#References)\n",
    "\n",
    "**Note**: this notebook is compatible with Statsmodels v0.12 and later. It will not work with Statsmodels v0.11 and earlier."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "import types\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import statsmodels.api as sm\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Brief overview\n",
    "\n",
    "The dynamic factor model considered in this notebook can be found in the `DynamicFactorMQ` class, which is a part of the time series analysis component (and in particular the state space models subcomponent) of Statsmodels. It can be accessed as follows:\n",
    "\n",
    "```python\n",
    "import statsmodels.api as sm\n",
    "model = sm.tsa.DynamicFactorMQ(...)\n",
    "```\n",
    "\n",
    "**Data**\n",
    "\n",
    "In this notebook, we'll use 127 monthly variables and 1 quarterly variable (GDP) from the [FRED-MD / FRED-QD dataset](https://research.stlouisfed.org/econ/mccracken/fred-databases/) (McCracken and Ng, 2016).\n",
    "\n",
    "**Statistical model**:\n",
    "\n",
    "The statistical model and the EM-algorithm used for parameter estimation are described in:\n",
    "\n",
    "- Bańbura and Modugno (2014), \"Maximum likelihood estimation of factor models on datasets with arbitrary pattern of missing data.\" ([Working paper](https://core.ac.uk/download/pdf/6684705.pdf), [Published](https://onlinelibrary.wiley.com/doi/full/10.1002/jae.2306?casa_token=tX0xS_49OXcAAAAA%3Aocw-egTRztTVg643NCHRCQUs_OGCPMTS78Qds4gk2nN6ViFjOMZYSDVip-0eeDwQCpvaTOTqjof5_wKI)), and\n",
    "- Bańbura et al. (2011), \"Nowcasting\" ([Working paper](https://core.ac.uk/download/pdf/6518537.pdf), [Handbook chapter](https://www.oxfordhandbooks.com/view/10.1093/oxfordhb/9780195398649.001.0001/oxfordhb-9780195398649-e-8))\n",
    "\n",
    "As in these papers, the basic specification starts from the typical \"static form\" of the dynamic factor model:\n",
    "\n",
    "$$\n",
    "\\begin{aligned}\n",
    "y_t & = \\Lambda f_t + \\epsilon_t \\\\\n",
    "f_t & = A_1 f_{t-1} + \\dots + A_p f_{t-p} + u_t\n",
    "\\end{aligned}\n",
    "$$\n",
    "\n",
    "The `DynamicFactorMQ` class allows for the generalizations of the model described in the references above, including:\n",
    "\n",
    "- Limiting blocks of factors to only load on certain observed variables\n",
    "- Monthly and quarterly mixed-frequency data, along the lines described by Mariano and Murasawa (2010)\n",
    "- Allowing autocorrelation (via AR(1) specifications) in the idiosyncratic disturbances $\\epsilon_t$\n",
    "- Missing entries in the observed variables $y_t$\n",
    "\n",
    "**Parameter estimation**\n",
    "\n",
    "When there are a large number of observed variables $y_t$, there can be hundreds or even thousands of parameters to estimate. In this situation, it can be extremely difficult for the typical quasi-Newton optimization methods (which are employed in many of the Statsmodels time series models) to find the parameters that maximize the likelihood function. Since this model is designed to support a large panel of observed variables, the EM-algorithm is instead used, since it can be more robust (see the above references for details).\n",
    "\n",
    "In particular, the model above is in state space form, and so the [state space library in Statsmodels](https://www.statsmodels.org/stable/statespace.html) can be used to apply the Kalman filter and smoother routines that are required for estimation.\n",
    "\n",
    "**Forecasting and interpolation**\n",
    "\n",
    "Because the model is in state space form, once the parameters are estimated, it is straightforward to produce forecasts of any of the observed variables $y_t$.\n",
    "\n",
    "In addition, for any missing entries in the observed variables $y_t$, it is also straightforward to produce estimates of those entries based on the factors extracted from the entire observed dataset (\"smoothed\" factor estimates). In a monthly/quarterly mixed frequency model, this can be used to interpolate monthly values for quarterly variables.\n",
    "\n",
    "**Nowcasting, updating forecasts, and computing the \"news\"**\n",
    "\n",
    "By including both monthly and quarterly variables, this model can be used to produce \"nowcasts\" of a quarterly variable before it is released, based on the monthly data for that quarter. For example, the advance estimate for the first quarter GDP is typically released in April, but this model could produce an estimate in March that was based on data through February.\n",
    "\n",
    "Many forecasting and nowcasting exercises are updated frequently, in \"real time\", and it is therefore important that one can easily add new datapoints as they come in. As these new datapoints provide new information, the model's forecast / nowcast will change with new data, and it is also important that one can easily decompose these changes into contributions from each updated series to changes in the forecast.  Both of these steps are supported by all state space models in Statsmodels – including the `DynamicFactorMQ` model – as we show below.\n",
    "\n",
    "**Other resources**\n",
    "\n",
    "The [New York Fed Staff Nowcast](https://www.newyorkfed.org/research/policy/nowcast.html) is an application of this same dynamic factor model and (EM algorithm) estimation method. Although they use a different dataset, and update their results weekly, their underlying framework is the same as that used in this notebook.\n",
    "\n",
    "For more details on the New York Fed Staff Nowcast model and results, see [Bok et al. (2018)](https://www.newyorkfed.org/research/staff_reports/sr830)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this notebook, we estimate a dynamic factor model on a large panel of economic data released at a monthly frequency, along with GDP, which is only released at a quarterly frequency. The monthly datasets that we'll be using come from [FRED-MD database](https://research.stlouisfed.org/econ/mccracken/fred-databases/) (McCracken and Ng, 2016), and we will take real GDP from the companion FRED-QD database.\n",
    "\n",
    "**Data vintage**\n",
    "\n",
    "The FRED-MD dataset was launched in January 2015, and vintages are available for each month since then. The FRED-QD dataset was fully launched in May 2018, and monthly vintages are available for each month since then. Our baseline exercise will use the February 2020 dataset (which includes data through reference month January 2020), and then we will examine how updated incoming data in the months of March - June influenced the model's forecast of real GDP growth in 2020Q2.\n",
    "\n",
    "**Data transformations**\n",
    "\n",
    "The assumptions of the dynamic factor model in this notebook require that the factors and observed variables are stationary. However, this dataset contains raw economic series that clearly violate that assumptions – for example, many of them show distinct trends. As is typical in these exercises, we therefore transform the variables to induce stationarity. In particular, the FRED-MD and FRED-QD datasets include suggested transformations (coded 1-7, which typically apply differences or percentage change transformations) which we apply.\n",
    "\n",
    "The exact details are in the `transform` function, below:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def transform(column, transforms):\n",
    "    transformation = transforms[column.name]\n",
    "    # For quarterly data like GDP, we will compute\n",
    "    # annualized percent changes\n",
    "    mult = 4 if column.index.freqstr[0] == 'Q' else 1\n",
    "    \n",
    "    # 1 => No transformation\n",
    "    if transformation == 1:\n",
    "        pass\n",
    "    # 2 => First difference\n",
    "    elif transformation == 2:\n",
    "        column = column.diff()\n",
    "    # 3 => Second difference\n",
    "    elif transformation == 3:\n",
    "        column = column.diff().diff()\n",
    "    # 4 => Log\n",
    "    elif transformation == 4:\n",
    "        column = np.log(column)\n",
    "    # 5 => Log first difference, multiplied by 100\n",
    "    #      (i.e. approximate percent change)\n",
    "    #      with optional multiplier for annualization\n",
    "    elif transformation == 5:\n",
    "        column = np.log(column).diff() * 100 * mult\n",
    "    # 6 => Log second difference, multiplied by 100\n",
    "    #      with optional multiplier for annualization\n",
    "    elif transformation == 6:\n",
    "        column = np.log(column).diff().diff() * 100 * mult\n",
    "    # 7 => Exact percent change, multiplied by 100\n",
    "    #      with optional annualization\n",
    "    elif transformation == 7:\n",
    "        column = ((column / column.shift(1))**mult - 1.0) * 100\n",
    "        \n",
    "    return column\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Outliers**\n",
    "\n",
    "Following McCracken and Ng (2016), we remove outliers (setting their value to missing), defined as observations that are more than 10 times the interquartile range from the series mean.\n",
    "\n",
    "However, in this exercise we are interested in \"nowcasting\" real GDP growth for 2020Q2, which was greatly affected by economic shutdowns stemming from the COVID-19 pandemic. During the first half of 2020, there are a number of series which include extreme observations, many of which would be excluded by this outlier test. Because these observations are likely to be informative about real GDP in 2020Q2, we only remove outliers for the period 1959-01 through 2019-12.\n",
    "\n",
    "The details of outlier removal are in the `remove_outliers` function, below:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def remove_outliers(dta):\n",
    "    # Compute the mean and interquartile range\n",
    "    mean = dta.mean()\n",
    "    iqr = dta.quantile([0.25, 0.75]).diff().T.iloc[:, 1]\n",
    "    \n",
    "    # Replace entries that are more than 10 times the IQR\n",
    "    # away from the mean with NaN (denotes a missing entry)\n",
    "    mask = np.abs(dta) > mean + 10 * iqr\n",
    "    treated = dta.copy()\n",
    "    treated[mask] = np.nan\n",
    "\n",
    "    return treated"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Loading the data**\n",
    "\n",
    "The `load_fredmd_data` function, below, performs the following actions, once for the FRED-MD dataset and once for the FRED-QD dataset:\n",
    "\n",
    "1. Based on the `vintage` argument, it downloads a particular vintage of these datasets from the base URL https://s3.amazonaws.com/files.fred.stlouisfed.org/fred-md into the `orig_[m|q]` variable.\n",
    "2. Extracts the column describing which transformation to apply into the `transform_[m|q]` (and, for the quarterly dataset, also extracts the column describing which factor an earlier paper assigned each variable to).\n",
    "3. Extracts the observation date (from the \"sasdate\" column) and uses it as the index of the dataset.\n",
    "4. Applies the transformations from step (2).\n",
    "5. Removes outliers for the period 1959-01 through 2019-12.\n",
    "\n",
    "Finally, these are collected into an easy-to-use object (the `SimpleNamespace` object) and returned."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_fredmd_data(vintage):\n",
    "    base_url = 'https://s3.amazonaws.com/files.fred.stlouisfed.org/fred-md'\n",
    "    \n",
    "    # - FRED-MD --------------------------------------------------------------\n",
    "    # 1. Download data\n",
    "    orig_m = (pd.read_csv(f'{base_url}/monthly/{vintage}.csv')\n",
    "                .dropna(how='all'))\n",
    "    \n",
    "    # 2. Extract transformation information\n",
    "    transform_m = orig_m.iloc[0, 1:]\n",
    "    orig_m = orig_m.iloc[1:]\n",
    "\n",
    "    # 3. Extract the date as an index\n",
    "    orig_m.index = pd.PeriodIndex(orig_m.sasdate.tolist(), freq='M')\n",
    "    orig_m.drop('sasdate', axis=1, inplace=True)\n",
    "\n",
    "    # 4. Apply the transformations\n",
    "    dta_m = orig_m.apply(transform, axis=0,\n",
    "                         transforms=transform_m)\n",
    "\n",
    "    # 5. Remove outliers (but not in 2020)\n",
    "    dta_m.loc[:'2019-12'] = remove_outliers(dta_m.loc[:'2019-12'])\n",
    "\n",
    "    # - FRED-QD --------------------------------------------------------------\n",
    "    # 1. Download data\n",
    "    orig_q = (pd.read_csv(f'{base_url}/quarterly/{vintage}.csv')\n",
    "                .dropna(how='all'))\n",
    "\n",
    "    # 2. Extract factors and transformation information\n",
    "    factors_q = orig_q.iloc[0, 1:]\n",
    "    transform_q = orig_q.iloc[1, 1:]\n",
    "    orig_q = orig_q.iloc[2:]\n",
    "\n",
    "    # 3. Extract the date as an index\n",
    "    orig_q.index = pd.PeriodIndex(orig_q.sasdate.tolist(), freq='Q')\n",
    "    orig_q.drop('sasdate', axis=1, inplace=True)\n",
    "\n",
    "    # 4. Apply the transformations\n",
    "    dta_q = orig_q.apply(transform, axis=0,\n",
    "                          transforms=transform_q)\n",
    "\n",
    "    # 5. Remove outliers (but not in 2020)\n",
    "    dta_q.loc[:'2019Q4'] = remove_outliers(dta_q.loc[:'2019Q4'])\n",
    "    \n",
    "    # - Output datasets ------------------------------------------------------\n",
    "    return types.SimpleNamespace(\n",
    "        orig_m=orig_m, orig_q=orig_q,\n",
    "        dta_m=dta_m, transform_m=transform_m,\n",
    "        dta_q=dta_q, transform_q=transform_q, factors_q=factors_q)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will call this `load_fredmd_data` function for each vintage from February 2020 through June 2020."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load the vintages of data from FRED\n",
    "dta = {date: load_fredmd_data(date)\n",
    "       for date in ['2020-02', '2020-03', '2020-04', '2020-05', '2020-06']}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "For vintage 2020-02, there are 128 series and 733 observations, over the period 1959-01 to 2020-01.\n"
     ]
    }
   ],
   "source": [
    "# Print some information about the base dataset\n",
    "n, k = dta['2020-02'].dta_m.shape\n",
    "start = dta['2020-02'].dta_m.index[0]\n",
    "end = dta['2020-02'].dta_m.index[-1]\n",
    "\n",
    "print(f'For vintage 2020-02, there are {k} series and {n} observations,'\n",
    "      f' over the period {start} to {end}.')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To see how the transformation and outlier removal works, here we plot three graphs of the RPI variable (\"Real Personal Income\") over the period 2000-01 - 2020-01:\n",
    "\n",
    "1. The original dataset (which is in Billions of Chained 2012 Dollars)\n",
    "2. The transformed data (RPI had a transformation code of \"5\", corresponding to log first difference)\n",
    "3. The transformed data, with outliers removed\n",
    "\n",
    "Notice that the large negative growth rate in January 2013 was deemed to be an outlier and so was replaced with a missing value (a `NaN` value).\n",
    "\n",
    "The [BEA release at the time](https://www.bea.gov/news/2013/personal-income-and-outlays-january-2013) noted that this was related to \"the effect of special factors, which boosted employee contributions for government social insurance in January [2013] and which had boosted wages and salaries and personal dividends in December [2012].\"."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x432 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "with sns.color_palette('deep'):\n",
    "    fig, axes = plt.subplots(3, figsize=(14, 6))\n",
    "\n",
    "    # Plot the raw data from the February 2020 vintage, for:\n",
    "    # \n",
    "    vintage = '2020-02'\n",
    "    variable = 'RPI'\n",
    "    start = '2000-01'\n",
    "    end = '2020-01'\n",
    "\n",
    "    # 1. Plot the original dataset, for 2000-01 through 2020-01\n",
    "    dta[vintage].orig_m.loc[start:end, variable].plot(ax=axes[0])\n",
    "    axes[0].set(title='Original data', xlim=('2000','2020'), ylabel='Billons of $')\n",
    "\n",
    "    # 2. Plot the transformed data, still including outliers\n",
    "    # (we only stored the transformation with outliers removed, so\n",
    "    # here we'll manually perform the transformation)\n",
    "    transformed = transform(dta[vintage].orig_m[variable],\n",
    "                            dta[vintage].transform_m)\n",
    "    transformed.loc[start:end].plot(ax=axes[1])\n",
    "    mean = transformed.mean()\n",
    "    iqr = transformed.quantile([0.25, 0.75]).diff().iloc[1]\n",
    "    axes[1].hlines([mean - 10 * iqr, mean + 10 * iqr],\n",
    "                   transformed.index[0], transformed.index[-1],\n",
    "                   linestyles='--', linewidth=1)\n",
    "    axes[1].set(title='Transformed data, with bands showing outliers cutoffs',\n",
    "                xlim=('2000','2020'), ylim=(mean - 15 * iqr, mean + 15 * iqr),\n",
    "                ylabel='Percent')\n",
    "    axes[1].annotate('Outlier', xy=('2013-01', transformed.loc['2013-01']),\n",
    "                     xytext=('2014-01', -5.3), textcoords='data',\n",
    "                     arrowprops=dict(arrowstyle=\"->\", connectionstyle=\"arc3\"),)\n",
    "\n",
    "    # 3. Plot the transformed data, with outliers removed (see missing value for 2013-01)\n",
    "    dta[vintage].dta_m.loc[start:end, 'RPI'].plot(ax=axes[2])\n",
    "    axes[2].set(title='Transformed data, with outliers removed',\n",
    "                xlim=('2000','2020'), ylabel='Percent')\n",
    "    axes[2].annotate('Missing value in place of outlier', xy=('2013-01', -1),\n",
    "                     xytext=('2014-01', -2), textcoords='data',\n",
    "                     arrowprops=dict(arrowstyle=\"->\", connectionstyle=\"arc3\"))\n",
    "    \n",
    "    fig.suptitle('Real Personal Income (RPI)',\n",
    "                 fontsize=12, fontweight=600)\n",
    "\n",
    "    fig.tight_layout(rect=[0, 0.00, 1, 0.95]);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Data definitions and details**\n",
    "\n",
    "In addition to providing the raw data, McCracken and Ng (2016) and the FRED-MD/FRED-QD dataset provide additional information in appendices about the variables in each dataset:\n",
    "\n",
    "- [FRED-MD Updated Appendix](https://s3.amazonaws.com/files.fred.stlouisfed.org/fred-md/Appendix_Tables_Update.pdf)\n",
    "- [FRED-QD Updated Appendix](https://s3.amazonaws.com/files.fred.stlouisfed.org/fred-md/FRED-QD_appendix.pdf)\n",
    "\n",
    "In particular, we're interested in:\n",
    "\n",
    "- The human-readable description (e.g. the description of the variable \"RPI\" is \"Real Personal Income\")\n",
    "- The grouping of the variable (e.g. \"RPI\" is part of the \"Output and income\" group, while \"FEDFUNDS\" is in the \"Interest and exchange rates\" group).\n",
    "\n",
    "The descriptions make it easier to understand the results, while the variable groupings can be useful in specifying the factor block structure. For example, we may want to have one or more \"Global\" factors that load on all variables while at the same time having one or more group-specific factors that only load on variables in a particular group.\n",
    "\n",
    "We extracted the information from the appendices above into CSV files, which we load here:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>group</th>\n",
       "      <th>id</th>\n",
       "      <th>tcode</th>\n",
       "      <th>fred</th>\n",
       "      <th>description</th>\n",
       "      <th>gsi</th>\n",
       "      <th>gsi:description</th>\n",
       "      <th>asterisk</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>fred</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>RPI</th>\n",
       "      <td>Output and Income</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>RPI</td>\n",
       "      <td>Real Personal Income</td>\n",
       "      <td>M_14386177</td>\n",
       "      <td>PI</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>W875RX1</th>\n",
       "      <td>Output and Income</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>W875RX1</td>\n",
       "      <td>Real personal income ex transfer receipts</td>\n",
       "      <td>M_145256755</td>\n",
       "      <td>PI less transfers</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>INDPRO</th>\n",
       "      <td>Output and Income</td>\n",
       "      <td>6</td>\n",
       "      <td>5</td>\n",
       "      <td>INDPRO</td>\n",
       "      <td>IP Index</td>\n",
       "      <td>M_116460980</td>\n",
       "      <td>IP: total</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>IPFPNSS</th>\n",
       "      <td>Output and Income</td>\n",
       "      <td>7</td>\n",
       "      <td>5</td>\n",
       "      <td>IPFPNSS</td>\n",
       "      <td>IP: Final Products and Nonindustrial Supplies</td>\n",
       "      <td>M_116460981</td>\n",
       "      <td>IP: products</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>IPFINAL</th>\n",
       "      <td>Output and Income</td>\n",
       "      <td>8</td>\n",
       "      <td>5</td>\n",
       "      <td>IPFINAL</td>\n",
       "      <td>IP: Final Products (Market Group)</td>\n",
       "      <td>M_116461268</td>\n",
       "      <td>IP: final prod</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     group  id  tcode     fred  \\\n",
       "fred                                             \n",
       "RPI      Output and Income   1      5      RPI   \n",
       "W875RX1  Output and Income   2      5  W875RX1   \n",
       "INDPRO   Output and Income   6      5   INDPRO   \n",
       "IPFPNSS  Output and Income   7      5  IPFPNSS   \n",
       "IPFINAL  Output and Income   8      5  IPFINAL   \n",
       "\n",
       "                                           description          gsi  \\\n",
       "fred                                                                  \n",
       "RPI                               Real Personal Income   M_14386177   \n",
       "W875RX1      Real personal income ex transfer receipts  M_145256755   \n",
       "INDPRO                                        IP Index  M_116460980   \n",
       "IPFPNSS  IP: Final Products and Nonindustrial Supplies  M_116460981   \n",
       "IPFINAL              IP: Final Products (Market Group)  M_116461268   \n",
       "\n",
       "           gsi:description  asterisk  \n",
       "fred                                  \n",
       "RPI                     PI       NaN  \n",
       "W875RX1  PI less transfers       NaN  \n",
       "INDPRO           IP: total       NaN  \n",
       "IPFPNSS       IP: products       NaN  \n",
       "IPFINAL     IP: final prod       NaN  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Definitions from the Appendix for FRED-MD variables\n",
    "defn_m = pd.read_csv('fredmd_definitions.csv')\n",
    "defn_m.index = defn_m.fred\n",
    "\n",
    "# Definitions from the Appendix for FRED-QD variables\n",
    "defn_q = pd.read_csv('fredqd_definitions.csv')\n",
    "defn_q.index = defn_q.fred\n",
    "\n",
    "# Example of the information in these files:\n",
    "defn_m.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To aid interpretation of the results, we'll replace the names of our dataset with the \"description\" field."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Replace the names of the columns in each monthly and quarterly dataset\n",
    "map_m = defn_m['description'].to_dict()\n",
    "map_q = defn_q['description'].to_dict()\n",
    "for date, value in dta.items():\n",
    "    value.orig_m.columns = value.orig_m.columns.map(map_m)\n",
    "    value.dta_m.columns = value.dta_m.columns.map(map_m)\n",
    "    value.orig_q.columns = value.orig_q.columns.map(map_q)\n",
    "    value.dta_q.columns = value.dta_q.columns.map(map_q)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Data groups**\n",
    "\n",
    "Below, we get the groups for each series from the definition files above, and then show how many of the series that we'll be using fall into each of the groups.\n",
    "\n",
    "We'll also re-order the series by group, to make it easier to interpret the results.\n",
    "\n",
    "Since we're including the quarterly real GDP variable in our analysis, we need to assign it to one of the groups in the monthly dataset. It fits best in the \"Output and income\" group."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th># series in group</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>group</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Output and Income</th>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Labor Market</th>\n",
       "      <td>32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Housing</th>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Consumption, orders, and inventories</th>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Money and credit</th>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Interest and exchange rates</th>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Prices</th>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Stock market</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                      # series in group\n",
       "group                                                  \n",
       "Output and Income                                    18\n",
       "Labor Market                                         32\n",
       "Housing                                              10\n",
       "Consumption, orders, and inventories                 14\n",
       "Money and credit                                     14\n",
       "Interest and exchange rates                          22\n",
       "Prices                                               21\n",
       "Stock market                                          5"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Get the mapping of variable id to group name, for monthly variables\n",
    "groups = defn_m[['description', 'group']].copy()\n",
    "\n",
    "# Re-order the variables according to the definition CSV file\n",
    "# (which is ordered by group)\n",
    "columns = [name for name in defn_m['description']\n",
    "           if name in dta['2020-02'].dta_m.columns]\n",
    "for date in dta.keys():\n",
    "    dta[date].dta_m = dta[date].dta_m.reindex(columns, axis=1)\n",
    "\n",
    "# Add real GDP (our quarterly variable) into the \"Output and Income\" group\n",
    "gdp_description = defn_q.loc['GDPC1', 'description']\n",
    "groups = groups.append({'description': gdp_description, 'group': 'Output and Income'},\n",
    "                       ignore_index=True)\n",
    "\n",
    "# Display the number of variables in each group\n",
    "(groups.groupby('group', sort=False)\n",
    "       .count()\n",
    "       .rename({'description': '# series in group'}, axis=1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model specification\n",
    "\n",
    "Now we need to specify all the details of the specific dynamic factor model that we want to estimate. In particular, we will choose:\n",
    "\n",
    "1. Factor specification, including: how many factors to include, which variables load on which factors, and the order of the (vector) autoregression that the factor evolves according to.\n",
    "\n",
    "2. Whether or not to standardize the dataset for estimation.\n",
    "\n",
    "3. Whether to model the idiosyncratic error terms as AR(1) processes or as iid."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Factor specification**\n",
    "\n",
    "While there are a number of ways to carefully identify the number of factors that one should use, here we will just choose an ad-hoc structure as follows:\n",
    "\n",
    "- Two global factors (i.e. factors that load on all variables) that jointly evolve according to a VAR(4)\n",
    "- One group-specific factor (i.e. factors that load only on variables in their group) for each of the 8 groups that were described above, with each evolving according to an AR(1)\n",
    "\n",
    "In the `DynamicFactorMQ` model, the basic factor structure is specified using the `factors` argument, which must be one of the following:\n",
    "\n",
    "- An integer, which can be used if one wants to specify only that number of global factors. This only allows a very simple factor specification.\n",
    "- A dictionary with keys equal to observed variable names and values equal to a list of factor names. Note that if this approach is used, the all observed variables must be included in this dictionary.\n",
    "\n",
    "Because we want to specify a complex factor structure, we need to take the later route. As an example of what an entry in the dictionary would look like, consider the \"Real personal income\" (previously the RPI) variable. It should load on the \"Global\" factor and on the \"Output and Income\" group factor:\n",
    "\n",
    "```python\n",
    "factors = {\n",
    "    'Real Personal Income': ['Global', 'Output and Income'],\n",
    "    ...\n",
    "}\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['Global', 'Output and Income']\n"
     ]
    }
   ],
   "source": [
    "# Construct the variable => list of factors dictionary\n",
    "factors = {row['description']: ['Global', row['group']]\n",
    "           for ix, row in groups.iterrows()}\n",
    "\n",
    "# Check that we have the desired output for \"Real personal income\"\n",
    "print(factors['Real Personal Income'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Factor multiplicities**\n",
    "\n",
    "You might notice that above we said that we wanted to have two global factors, while in the `factors` dictionary we just specified, we only put in \"Global\" once. This is because we can use the `factor_multiplicities` argument to more conveniently set up multiple factors that evolve together jointly (in this example, it wouldn't have been too hard to specify something like `['Global.1', 'Global.2', 'Output and Income']`, but this can get tedious quickly).\n",
    "\n",
    "The `factor_multiplicities` argument defaults to `1`, but if it is specified then it must be one of the following:\n",
    "\n",
    "- An integer, which can be used if one wants to specify that every factor has the same multiplicity\n",
    "- A dictionary with keys equal to factor names (from the `factors` argument) and values equal to an integer. Note that the default for each factor is 1, so you only need to include in this dictionary factors that have multiplicity greater than 1.\n",
    "\n",
    "Here, we want all of the group factors to be univariate, while we want a bivariate set of global factors. Therefore, we only need to specify the `{'Global': 2}` part, while the rest will be assumed to have multiplicity 1 by default."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "factor_multiplicities = {'Global': 2}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Factor orders**\n",
    "\n",
    "Finally, we need to specify the lag order of the (vector) autoregressions that govern the dynamics of the factors. This is done via the `factor_orders` argument.\n",
    "\n",
    "The `factor_orders` argument defaults to `1`, but if it is specified then it must be one of the following:\n",
    "\n",
    "- An integer, which can be used if one wants to specify the same order of (vector) autoregression for each factor.\n",
    "- A dictionary with keys equal to factor names (from the `factors` argument) or tuples of factor names and values equal to an integer. Note that the default for each factor is 1, so you only need to include in this dictionary factors that have order greater than 1.\n",
    "\n",
    "The reason that the dictionary keys can be tuples of factor names is that this syntax allows you to specify \"blocks\" of factors that evolve jointly according to a vector autoregressive process rather than individually according to univariate autoregressions. Note that the default for each factor is a univariate autoregression of order 1, so we only need to include in this dictionary factors or blocks of factors that differ from that assumption.\n",
    "\n",
    "For example, if we had:\n",
    "\n",
    "```python\n",
    "factor_orders = {\n",
    "    'Output and Income': 2\n",
    "}\n",
    "```\n",
    "\n",
    "Then we would have:\n",
    "\n",
    "- All group-specific factors except \"Output and Income\" follow univarate AR(1) processes\n",
    "- The \"Output and Income\" group-specific factor follows a univarate AR(2) process\n",
    "- The two \"Global\" factors jointly follow a VAR(1) process (this is because multiple factor defined by a multiplicity greater than one are automatically assumed to evolve jointly)\n",
    "\n",
    "Alternatively, if we had:\n",
    "\n",
    "```python\n",
    "factor_orders = {\n",
    "    ('Output and Income', 'Labor Market'): 1\n",
    "    'Global': 2\n",
    "}\n",
    "```\n",
    "\n",
    "Then we would have:\n",
    "\n",
    "- All group-specific factors except \"Output and Income\" and \"Labor Market\" follow univarate AR(1) processes\n",
    "- The \"Output and Income\" and \"Labor Market\" group-specific factors joinlty follow a univarate VAR(1) process\n",
    "- The two \"Global\" factors jointly follow a VAR(2) process (this is again because multiple factor defined by a multiplicity greater than one are automatically assumed to evolve jointly)\n",
    "\n",
    "In this case, we only need to specify that the \"Global\" factors evolve according to a VAR(4), which can be done with:\n",
    "\n",
    "```python\n",
    "factor_orders = {'Global': 4}\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "factor_orders = {'Global': 4}"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Creating the model**\n",
    "\n",
    "Given the factor specification, above, we can finish the model specification and create the model object.\n",
    "\n",
    "The `DynamicFactorMQ` model class has the following primary arguments:\n",
    "\n",
    "1. `endog` and `endog_quarterly`\n",
    "\n",
    "   These arguments are used to pass the observed variables to the model. There are two ways to provide the data:\n",
    "   \n",
    "   1. If you are specifying a monthly / quarterly mixed frequency model, then you would pass the monthly data to `endog` and the quarterly data to the keyword argument `endog_quarterly`. This is what we have done below.\n",
    "   2. If you are specifying any other kind of model, then you simply pass all of your observed data to the `endog` variable and you do not include the `endog_quarterly` argument. In this case, the `endog` data does not need to be monthly - it can be any frequency (or no frequency).\n",
    "\n",
    "\n",
    "2. `factors`, `factor_orders`, and `factor_multiplicities`\n",
    "\n",
    "   These arguments were described above.\n",
    "\n",
    "\n",
    "3. `idiosyncratic_ar1`\n",
    "\n",
    "   As noted in the \"Brief Overview\" section, above, the `DynamicFactorMQ` model allows the idiosyncratic disturbance terms to be modeled as independent AR(1) processes or as iid variables. The default is `idiosyncratic_ar1=True`, which can be useful in modeling some of the idiosyncratic serial correlation, for example for forecasting.\n",
    "\n",
    "\n",
    "4. `standardize`\n",
    "\n",
    "   Although we earlier transformed all of the variables to be stationary, they will still fluctuate around different means and they may have very different scales. This can make it difficult to fit the model. In most applications, therefore, the variables are standardized by subtracting the mean and dividing by the standard deviation. This is the default, so we do not set this argument below.\n",
    "\n",
    "   It is recommended for users to use this argument rather than standardizing the variables themselves. This is because if the `standardize` argument is used, then the model will automatically reverse the standardization for all post-estimation results like prediction, forecasting, and computation of the impacts of the \"news\". This means that these results can be directly compared to the input data.\n",
    "   \n",
    "**Note**: to generate the best nowcasts for GDP growth in 2020Q2, we will restrict the sample to start in 2000-01 rather than in 1960 (for example, to guard against the possibility of structural changes in underlying parameters)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "# Get the baseline monthly and quarterly datasets\n",
    "start = '2000'\n",
    "endog_m = dta['2020-02'].dta_m.loc[start:, :]\n",
    "gdp_description = defn_q.loc['GDPC1', 'description']\n",
    "endog_q = dta['2020-02'].dta_q.loc[start:, [gdp_description]]\n",
    "\n",
    "# Construct the dynamic factor model\n",
    "model = sm.tsa.DynamicFactorMQ(\n",
    "    endog_m, endog_quarterly=endog_q,\n",
    "    factors=factors, factor_orders=factor_orders,\n",
    "    factor_multiplicities=factor_multiplicities)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Model summary**\n",
    "\n",
    "Because these models can be somewhat complex to set up, it can be useful to check the results of the model's `summary` method. This method produces three tables.\n",
    "\n",
    "1. **Model specification**: the first table shows general information about the model selected, the sample, factor setup, and other options.\n",
    "\n",
    "2. **Observed variables / factor loadings**: the second table shows which factors load on which observed variables. This table should be checked to make sure that the `factors` and `factor_multiplicities` arguments were specified as desired.\n",
    "\n",
    "3. **Factor block orders**: the last table shows the blocks of factors (the factors within each block evolve jointly, while between blocks the factors are independent) and the order of the (vector) autoregression. This table should be checked to make sure that the `factor_orders` argument was specified as desired.\n",
    "\n",
    "Note that by default, the names of the observed variables are truncated to prevent tables from overflowing. The length before truncation can be specified by changing the value of the `truncate_endog_names` argument."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>Model Specification: Dynamic Factor Model</caption>\n",
       "<tr>\n",
       "  <th>Model:</th>    <td>Dynamic Factor Model</td>   <th>  # of monthly variables:</th>      <td>128</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>        <td>+ 10 factors in 9 blocks</td> <th>  # of quarterly variables:</th>     <td>1</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>         <td>+ Mixed frequency (M/Q)</td> <th>  # of factor blocks:</th>           <td>9</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>          <td>+ AR(1) idiosyncratic</td>  <th>  Idiosyncratic disturbances:</th> <td>AR(1)</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Sample:</th>          <td>2000-01</td>         <th>  Standardize variables:</th>      <td>True</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                <td>- 2020-01</td>        <th>                     </th>           <td> </td>  \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Observed variables / factor loadings</caption>\n",
       "<tr>\n",
       "         <th>Dep. variable</th>        <th>Global.1</th> <th>Global.2</th> <th>Output and Income</th> <th>Labor Market</th> <th>Housing</th> <th>Consumption, orders, and inventories</th> <th>Money and credit</th> <th>Interest and exchange rates</th> <th>Prices</th> <th>Stock market</th>\n",
       "</tr>\n",
       "<tr>\n",
       "     <td>Real Personal Income</td>       <td>X    </td>    <td>X    </td>      <td>X        </td>        <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Real personal income ex ...</td>   <td>X    </td>    <td>X    </td>      <td>X        </td>        <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "           <td>IP Index</td>             <td>X    </td>    <td>X    </td>      <td>X        </td>        <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>IP: Final Products and N...</td>   <td>X    </td>    <td>X    </td>      <td>X        </td>        <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>IP: Final Products (Mark...</td>   <td>X    </td>    <td>X    </td>      <td>X        </td>        <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "      <td>IP: Consumer Goods</td>        <td>X    </td>    <td>X    </td>      <td>X        </td>        <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>IP: Durable Consumer Goo...</td>   <td>X    </td>    <td>X    </td>      <td>X        </td>        <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>IP: Nondurable Consumer ...</td>   <td>X    </td>    <td>X    </td>      <td>X        </td>        <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "    <td>IP: Business Equipment</td>      <td>X    </td>    <td>X    </td>      <td>X        </td>        <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "         <td>IP: Materials</td>          <td>X    </td>    <td>X    </td>      <td>X        </td>        <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "     <td>IP: Durable Materials</td>      <td>X    </td>    <td>X    </td>      <td>X        </td>        <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "   <td>IP: Nondurable Materials</td>     <td>X    </td>    <td>X    </td>      <td>X        </td>        <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "    <td>IP: Manufacturing (SIC)</td>     <td>X    </td>    <td>X    </td>      <td>X        </td>        <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>IP: Residential Utilitie...</td>   <td>X    </td>    <td>X    </td>      <td>X        </td>        <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "           <td>IP: Fuels</td>            <td>X    </td>    <td>X    </td>      <td>X        </td>        <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Capacity Utilization: Ma...</td>   <td>X    </td>    <td>X    </td>      <td>X        </td>        <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Help-Wanted Index for Un...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Ratio of Help Wanted/No....</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "     <td>Civilian Labor Force</td>       <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "      <td>Civilian Employment</td>       <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Civilian Unemployment Ra...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Average Duration of Unem...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Civilians Unemployed - L...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Civilians Unemployed for...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Civilians Unemployed - 1...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Civilians Unemployed for...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Civilians Unemployed for...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "        <td>Initial Claims</td>          <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Total non...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Goods-Pro...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Mining an...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Construct...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Manufactu...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Durable g...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Nondurabl...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Service-P...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Trade, Tr...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Wholesale...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Retail Tr...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Financial...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Governmen...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Avg Weekly Hours : Goods...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Avg Weekly Overtime Hour...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Avg Weekly Hours : Manuf...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Avg Hourly Earnings : Go...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Avg Hourly Earnings : Co...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Avg Hourly Earnings : Ma...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>X      </td>     <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Housing Starts: Total Ne...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>     <td>X   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Housing Starts, Northeas...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>     <td>X   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "    <td>Housing Starts, Midwest</td>     <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>     <td>X   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "     <td>Housing Starts, South</td>      <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>     <td>X   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "     <td>Housing Starts, West</td>       <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>     <td>X   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>New Private Housing Perm...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>     <td>X   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>New Private Housing Perm...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>     <td>X   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>New Private Housing Perm...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>     <td>X   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>New Private Housing Perm...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>     <td>X   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>New Private Housing Perm...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>     <td>X   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Real personal consumptio...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>X                  </td>             <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Real Manu. and Trade Ind...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>X                  </td>             <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Retail and Food Services...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>X                  </td>             <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>New Orders for Consumer ...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>X                  </td>             <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>New Orders for Durable G...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>X                  </td>             <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>New Orders for Nondefens...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>X                  </td>             <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Unfilled Orders for Dura...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>X                  </td>             <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Total Business Inventori...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>X                  </td>             <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Total Business: Inventor...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>X                  </td>             <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "   <td>Consumer Sentiment Index</td>     <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>X                  </td>             <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "        <td>M1 Money Stock</td>          <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>X        </td>           <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "        <td>M2 Money Stock</td>          <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>X        </td>           <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "      <td>Real M2 Money Stock</td>       <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>X        </td>           <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "         <td>Monetary Base</td>          <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>X        </td>           <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Total Reserves of Deposi...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>X        </td>           <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Reserves Of Depository I...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>X        </td>           <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Commercial and Industria...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>X        </td>           <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Real Estate Loans at All...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>X        </td>           <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Total Nonrevolving Credi...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>X        </td>           <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Nonrevolving consumer cr...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>X        </td>           <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "        <td>MZM Money Stock</td>         <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>X        </td>           <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Consumer Motor Vehicle L...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>X        </td>           <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Total Consumer Loans and...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>X        </td>           <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Securities in Bank Credi...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>X        </td>           <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Effective Federal Funds ...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>           <td>X             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>3-Month AA Financial Com...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>           <td>X             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "    <td>3-Month Treasury Bill:</td>      <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>           <td>X             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "    <td>6-Month Treasury Bill:</td>      <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>           <td>X             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "     <td>1-Year Treasury Rate</td>       <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>           <td>X             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "     <td>5-Year Treasury Rate</td>       <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>           <td>X             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "     <td>10-Year Treasury Rate</td>      <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>           <td>X             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Moody’s Seasoned Aaa Cor...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>           <td>X             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Moody’s Seasoned Baa Cor...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>           <td>X             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>3-Month Commercial Paper...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>           <td>X             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>3-Month Treasury C Minus...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>           <td>X             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>6-Month Treasury C Minus...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>           <td>X             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>1-Year Treasury C Minus ...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>           <td>X             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>5-Year Treasury C Minus ...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>           <td>X             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>10-Year Treasury C Minus...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>           <td>X             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Moody’s Aaa Corporate Bo...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>           <td>X             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Moody’s Baa Corporate Bo...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>           <td>X             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Trade Weighted U.S. Doll...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>           <td>X             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Switzerland / U.S. Forei...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>           <td>X             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Japan / U.S. Foreign Exc...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>           <td>X             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>U.S. / U.K. Foreign Exch...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>           <td>X             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Canada / U.S. Foreign Ex...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>           <td>X             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "      <td>PPI: Finished Goods</td>       <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>         <td>X   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>PPI: Finished Consumer G...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>         <td>X   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>PPI: Intermediate Materi...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>         <td>X   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "     <td>PPI: Crude Materials</td>       <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>         <td>X   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Crude Oil, spliced WTI a...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>         <td>X   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>PPI: Metals and metal pr...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>         <td>X   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "        <td>CPI : All Items</td>         <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>         <td>X   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "         <td>CPI : Apparel</td>          <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>         <td>X   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "     <td>CPI : Transportation</td>       <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>         <td>X   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "      <td>CPI : Medical Care</td>        <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>         <td>X   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "       <td>CPI : Commodities</td>        <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>         <td>X   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "        <td>CPI : Durables</td>          <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>         <td>X   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "        <td>CPI : Services</td>          <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>         <td>X   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>CPI : All Items Less Foo...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>         <td>X   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>CPI : All items less she...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>         <td>X   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>CPI : All items less med...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>         <td>X   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Personal Cons. Expend.: ...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>         <td>X   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Personal Cons. Exp: Dura...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>         <td>X   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Personal Cons. Exp: Nond...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>         <td>X   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Personal Cons. Exp: Serv...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>         <td>X   </td>     <td>      </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>S&P’s Common Stock Price...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>X      </td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>S&P’s Common Stock Price...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>X      </td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>S&P’s Composite Common S...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>X      </td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>S&P’s Composite Common S...</td>   <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>X      </td>  \n",
       "</tr>\n",
       "<tr>\n",
       "              <td>VXO</td>               <td>X    </td>    <td>X    </td>      <td>        </td>         <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>X      </td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Real Gross Domestic Prod...</td>   <td>X    </td>    <td>X    </td>      <td>X        </td>        <td>      </td>      <td>   </td>            <td>                  </td>              <td>        </td>            <td>             </td>          <td>   </td>     <td>      </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Factor blocks:</caption>\n",
       "<tr>\n",
       "                  <th>block</th>                <th>order</th>\n",
       "</tr>\n",
       "<tr>\n",
       "           <td>Global.1, Global.2</td>            <td>4</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "            <td>Output and Income</td>            <td>1</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "              <td>Labor Market</td>               <td>1</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "                 <td>Housing</td>                 <td>1</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Consumption, orders, and inventories</td>   <td>1</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "            <td>Money and credit</td>             <td>1</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "       <td>Interest and exchange rates</td>       <td>1</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "                 <td>Prices</td>                  <td>1</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "              <td>Stock market</td>               <td>1</td>  \n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                         Model Specification: Dynamic Factor Model                          \n",
       "============================================================================================\n",
       "Model:                 Dynamic Factor Model   # of monthly variables:                    128\n",
       "                   + 10 factors in 9 blocks   # of quarterly variables:                    1\n",
       "                    + Mixed frequency (M/Q)   # of factor blocks:                          9\n",
       "                      + AR(1) idiosyncratic   Idiosyncratic disturbances:              AR(1)\n",
       "Sample:                             2000-01   Standardize variables:                    True\n",
       "                                  - 2020-01                                                 \n",
       "                                                                                 Observed variables / factor loadings                                                                                 \n",
       "======================================================================================================================================================================================================\n",
       "              Dep. variable   Global.1   Global.2 Output and Income Labor Market     Housing Consumption, orders, and inventories Money and credit Interest and exchange rates     Prices Stock market\n",
       "------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\n",
       "       Real Personal Income      X          X             X                                                                                                                                           \n",
       "Real personal income ex ...      X          X             X                                                                                                                                           \n",
       "                   IP Index      X          X             X                                                                                                                                           \n",
       "IP: Final Products and N...      X          X             X                                                                                                                                           \n",
       "IP: Final Products (Mark...      X          X             X                                                                                                                                           \n",
       "         IP: Consumer Goods      X          X             X                                                                                                                                           \n",
       "IP: Durable Consumer Goo...      X          X             X                                                                                                                                           \n",
       "IP: Nondurable Consumer ...      X          X             X                                                                                                                                           \n",
       "     IP: Business Equipment      X          X             X                                                                                                                                           \n",
       "              IP: Materials      X          X             X                                                                                                                                           \n",
       "      IP: Durable Materials      X          X             X                                                                                                                                           \n",
       "   IP: Nondurable Materials      X          X             X                                                                                                                                           \n",
       "    IP: Manufacturing (SIC)      X          X             X                                                                                                                                           \n",
       "IP: Residential Utilitie...      X          X             X                                                                                                                                           \n",
       "                  IP: Fuels      X          X             X                                                                                                                                           \n",
       "Capacity Utilization: Ma...      X          X             X                                                                                                                                           \n",
       "Help-Wanted Index for Un...      X          X                            X                                                                                                                            \n",
       "Ratio of Help Wanted/No....      X          X                            X                                                                                                                            \n",
       "       Civilian Labor Force      X          X                            X                                                                                                                            \n",
       "        Civilian Employment      X          X                            X                                                                                                                            \n",
       "Civilian Unemployment Ra...      X          X                            X                                                                                                                            \n",
       "Average Duration of Unem...      X          X                            X                                                                                                                            \n",
       "Civilians Unemployed - L...      X          X                            X                                                                                                                            \n",
       "Civilians Unemployed for...      X          X                            X                                                                                                                            \n",
       "Civilians Unemployed - 1...      X          X                            X                                                                                                                            \n",
       "Civilians Unemployed for...      X          X                            X                                                                                                                            \n",
       "Civilians Unemployed for...      X          X                            X                                                                                                                            \n",
       "             Initial Claims      X          X                            X                                                                                                                            \n",
       "All Employees: Total non...      X          X                            X                                                                                                                            \n",
       "All Employees: Goods-Pro...      X          X                            X                                                                                                                            \n",
       "All Employees: Mining an...      X          X                            X                                                                                                                            \n",
       "All Employees: Construct...      X          X                            X                                                                                                                            \n",
       "All Employees: Manufactu...      X          X                            X                                                                                                                            \n",
       "All Employees: Durable g...      X          X                            X                                                                                                                            \n",
       "All Employees: Nondurabl...      X          X                            X                                                                                                                            \n",
       "All Employees: Service-P...      X          X                            X                                                                                                                            \n",
       "All Employees: Trade, Tr...      X          X                            X                                                                                                                            \n",
       "All Employees: Wholesale...      X          X                            X                                                                                                                            \n",
       "All Employees: Retail Tr...      X          X                            X                                                                                                                            \n",
       "All Employees: Financial...      X          X                            X                                                                                                                            \n",
       "All Employees: Governmen...      X          X                            X                                                                                                                            \n",
       "Avg Weekly Hours : Goods...      X          X                            X                                                                                                                            \n",
       "Avg Weekly Overtime Hour...      X          X                            X                                                                                                                            \n",
       "Avg Weekly Hours : Manuf...      X          X                            X                                                                                                                            \n",
       "Avg Hourly Earnings : Go...      X          X                            X                                                                                                                            \n",
       "Avg Hourly Earnings : Co...      X          X                            X                                                                                                                            \n",
       "Avg Hourly Earnings : Ma...      X          X                            X                                                                                                                            \n",
       "Housing Starts: Total Ne...      X          X                                           X                                                                                                             \n",
       "Housing Starts, Northeas...      X          X                                           X                                                                                                             \n",
       "    Housing Starts, Midwest      X          X                                           X                                                                                                             \n",
       "      Housing Starts, South      X          X                                           X                                                                                                             \n",
       "       Housing Starts, West      X          X                                           X                                                                                                             \n",
       "New Private Housing Perm...      X          X                                           X                                                                                                             \n",
       "New Private Housing Perm...      X          X                                           X                                                                                                             \n",
       "New Private Housing Perm...      X          X                                           X                                                                                                             \n",
       "New Private Housing Perm...      X          X                                           X                                                                                                             \n",
       "New Private Housing Perm...      X          X                                           X                                                                                                             \n",
       "Real personal consumptio...      X          X                                                                 X                                                                                       \n",
       "Real Manu. and Trade Ind...      X          X                                                                 X                                                                                       \n",
       "Retail and Food Services...      X          X                                                                 X                                                                                       \n",
       "New Orders for Consumer ...      X          X                                                                 X                                                                                       \n",
       "New Orders for Durable G...      X          X                                                                 X                                                                                       \n",
       "New Orders for Nondefens...      X          X                                                                 X                                                                                       \n",
       "Unfilled Orders for Dura...      X          X                                                                 X                                                                                       \n",
       "Total Business Inventori...      X          X                                                                 X                                                                                       \n",
       "Total Business: Inventor...      X          X                                                                 X                                                                                       \n",
       "   Consumer Sentiment Index      X          X                                                                 X                                                                                       \n",
       "             M1 Money Stock      X          X                                                                                            X                                                            \n",
       "             M2 Money Stock      X          X                                                                                            X                                                            \n",
       "        Real M2 Money Stock      X          X                                                                                            X                                                            \n",
       "              Monetary Base      X          X                                                                                            X                                                            \n",
       "Total Reserves of Deposi...      X          X                                                                                            X                                                            \n",
       "Reserves Of Depository I...      X          X                                                                                            X                                                            \n",
       "Commercial and Industria...      X          X                                                                                            X                                                            \n",
       "Real Estate Loans at All...      X          X                                                                                            X                                                            \n",
       "Total Nonrevolving Credi...      X          X                                                                                            X                                                            \n",
       "Nonrevolving consumer cr...      X          X                                                                                            X                                                            \n",
       "            MZM Money Stock      X          X                                                                                            X                                                            \n",
       "Consumer Motor Vehicle L...      X          X                                                                                            X                                                            \n",
       "Total Consumer Loans and...      X          X                                                                                            X                                                            \n",
       "Securities in Bank Credi...      X          X                                                                                            X                                                            \n",
       "Effective Federal Funds ...      X          X                                                                                                                   X                                     \n",
       "3-Month AA Financial Com...      X          X                                                                                                                   X                                     \n",
       "     3-Month Treasury Bill:      X          X                                                                                                                   X                                     \n",
       "     6-Month Treasury Bill:      X          X                                                                                                                   X                                     \n",
       "       1-Year Treasury Rate      X          X                                                                                                                   X                                     \n",
       "       5-Year Treasury Rate      X          X                                                                                                                   X                                     \n",
       "      10-Year Treasury Rate      X          X                                                                                                                   X                                     \n",
       "Moody’s Seasoned Aaa Cor...      X          X                                                                                                                   X                                     \n",
       "Moody’s Seasoned Baa Cor...      X          X                                                                                                                   X                                     \n",
       "3-Month Commercial Paper...      X          X                                                                                                                   X                                     \n",
       "3-Month Treasury C Minus...      X          X                                                                                                                   X                                     \n",
       "6-Month Treasury C Minus...      X          X                                                                                                                   X                                     \n",
       "1-Year Treasury C Minus ...      X          X                                                                                                                   X                                     \n",
       "5-Year Treasury C Minus ...      X          X                                                                                                                   X                                     \n",
       "10-Year Treasury C Minus...      X          X                                                                                                                   X                                     \n",
       "Moody’s Aaa Corporate Bo...      X          X                                                                                                                   X                                     \n",
       "Moody’s Baa Corporate Bo...      X          X                                                                                                                   X                                     \n",
       "Trade Weighted U.S. Doll...      X          X                                                                                                                   X                                     \n",
       "Switzerland / U.S. Forei...      X          X                                                                                                                   X                                     \n",
       "Japan / U.S. Foreign Exc...      X          X                                                                                                                   X                                     \n",
       "U.S. / U.K. Foreign Exch...      X          X                                                                                                                   X                                     \n",
       "Canada / U.S. Foreign Ex...      X          X                                                                                                                   X                                     \n",
       "        PPI: Finished Goods      X          X                                                                                                                                        X                \n",
       "PPI: Finished Consumer G...      X          X                                                                                                                                        X                \n",
       "PPI: Intermediate Materi...      X          X                                                                                                                                        X                \n",
       "       PPI: Crude Materials      X          X                                                                                                                                        X                \n",
       "Crude Oil, spliced WTI a...      X          X                                                                                                                                        X                \n",
       "PPI: Metals and metal pr...      X          X                                                                                                                                        X                \n",
       "            CPI : All Items      X          X                                                                                                                                        X                \n",
       "              CPI : Apparel      X          X                                                                                                                                        X                \n",
       "       CPI : Transportation      X          X                                                                                                                                        X                \n",
       "         CPI : Medical Care      X          X                                                                                                                                        X                \n",
       "          CPI : Commodities      X          X                                                                                                                                        X                \n",
       "             CPI : Durables      X          X                                                                                                                                        X                \n",
       "             CPI : Services      X          X                                                                                                                                        X                \n",
       "CPI : All Items Less Foo...      X          X                                                                                                                                        X                \n",
       "CPI : All items less she...      X          X                                                                                                                                        X                \n",
       "CPI : All items less med...      X          X                                                                                                                                        X                \n",
       "Personal Cons. Expend.: ...      X          X                                                                                                                                        X                \n",
       "Personal Cons. Exp: Dura...      X          X                                                                                                                                        X                \n",
       "Personal Cons. Exp: Nond...      X          X                                                                                                                                        X                \n",
       "Personal Cons. Exp: Serv...      X          X                                                                                                                                        X                \n",
       "S&P’s Common Stock Price...      X          X                                                                                                                                                  X      \n",
       "S&P’s Common Stock Price...      X          X                                                                                                                                                  X      \n",
       "S&P’s Composite Common S...      X          X                                                                                                                                                  X      \n",
       "S&P’s Composite Common S...      X          X                                                                                                                                                  X      \n",
       "                        VXO      X          X                                                                                                                                                  X      \n",
       "Real Gross Domestic Prod...      X          X             X                                                                                                                                           \n",
       "                 Factor blocks:                \n",
       "===============================================\n",
       "                               block      order\n",
       "-----------------------------------------------\n",
       "                  Global.1, Global.2          4\n",
       "                   Output and Income          1\n",
       "                        Labor Market          1\n",
       "                             Housing          1\n",
       "Consumption, orders, and inventories          1\n",
       "                    Money and credit          1\n",
       "         Interest and exchange rates          1\n",
       "                              Prices          1\n",
       "                        Stock market          1\n",
       "===============================================\n",
       "\"\"\""
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model fitting / parameter estimation\n",
    "\n",
    "With the model constructed as shown above, the model can be fit / the parameters can be estimated via the `fit` method. This method does not affect the `model` object that we created before, but instead returns a new `results` object that contains the estimates of the parameters and the state vector, and also allows forecasting and computation of the \"news\" when updated data arrives.\n",
    "\n",
    "The default method for parameter estimation in the `DynamicFactorMQ` class is maximum likelihood via the EM algorithm.\n",
    "\n",
    "**Note**: by default, the `fit` method does not show any output. Here we use the `disp=10` method to print details of every 10th iteration of the EM algorithm, to track its progress."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "EM start iterations, llf=-28556\n",
      "EM iteration 10, llf=-25703, convergence criterion=0.00045107\n",
      "EM iteration 20, llf=-25673, convergence criterion=3.1194e-05\n",
      "EM iteration 30, llf=-25669, convergence criterion=8.8719e-06\n",
      "EM iteration 40, llf=-25668, convergence criterion=3.7567e-06\n",
      "EM iteration 50, llf=-25667, convergence criterion=2.0379e-06\n",
      "EM iteration 60, llf=-25667, convergence criterion=1.3147e-06\n",
      "EM converged at iteration 69, llf=-25666, convergence criterion=9.7195e-07 < tolerance=1e-06\n"
     ]
    }
   ],
   "source": [
    "results = model.fit(disp=10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A summary of the model results including the estimated parameters can be produced using the `summary` method. This method produces three or four sets of tables. To save space in the output, here we are using `display_diagnostics=False` to hide the table showing residual diagnostics for the observed variables, so that only three sets of tables are shown.\n",
    "\n",
    "1. **Model specification**: the first table shows general information about the model selected, the sample, and summary values like the sample log likelihood and information criteria.\n",
    "\n",
    "2. **Observation equation**: the second table shows the estimated factor loadings for each observed variable / factor combination, or a dot (.) when a given variable does not load on a given factor. The last one or two columns show parameter estimates related to the idiosyncratic disturbance. In the `idiosyncratic_ar1=True` case, there are two columns at the end, one showing the estimated autoregressive coefficient and one showing the estimated variance of the disturbance innovation.\n",
    "\n",
    "3. **Factor block transition equations**: the next set of tables show the estimated (vector) autoregressive transition equations for each factor block. The first columns show the autoregressive coefficients, and the final columns show the error variance or covariance matrix.\n",
    "\n",
    "4. **(Optional) Residual diagnostics**: the last table (optional) shows three diagnostics from the (standardized) residuals associated with each observed variable.\n",
    "\n",
    "   1. First, the Ljung-Box test statistic for the first lag is shown to test for serial correlation.\n",
    "   2. Second, a test for heteroskedasticity.\n",
    "   3. Third, the Jarque-Bera test for normality.\n",
    "   \n",
    "   The default for dynamic factor models is not to show these diagnostics, which can be difficult to intepret for quarterly variables. However, the table can be shown by passing the `display_diagnostics=True` argument to the `summary` method.\n",
    "\n",
    "Note that by default, the names of the observed variables are truncated to prevent tables from overflowing. The length before truncation can be specified by changing the value of the `truncate_endog_names` argument."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>Dynamic Factor Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>   <td>\"Real Personal Income\", and 128 more</td> <th>  No. Observations:  </th>     <td>241</td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>Dynamic Factor Model</td>         <th>  Log Likelihood     </th> <td>-25666.313</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                       <td>+ 10 factors in 9 blocks</td>       <th>  AIC                </th>  <td>52692.627</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                        <td>+ Mixed frequency (M/Q)</td>       <th>  BIC                </th>  <td>55062.289</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                         <td>+ AR(1) idiosyncratic</td>        <th>  HQIC               </th>  <td>53647.320</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>                      <td>Fri, 31 Jul 2020</td>           <th>  EM Iterations      </th>      <td>6</td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                          <td>01:28:24</td>               <th>                     </th>      <td> </td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Sample:</th>                       <td>01-31-2000</td>              <th>                     </th>      <td> </td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th></th>                             <td>- 01-31-2020</td>             <th>                     </th>      <td> </td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>             <td>Not computed</td>             <th>                     </th>      <td> </td>    \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Observation equation:</caption>\n",
       "<tr>\n",
       "       <th>Factor loadings:</th>       <th>Global.1</th> <th>Global.2</th> <th>Output and Income</th> <th>Labor Market</th> <th>Housing</th> <th>Consumption, orders, and inventories</th> <th>Money and credit</th> <th>Interest and exchange rates</th> <th>Prices</th> <th>Stock market</th> <th>   idiosyncratic: AR(1)</th> <th>var.</th>\n",
       "</tr>\n",
       "<tr>\n",
       "     <td>Real Personal Income</td>       <td>-0.07</td>    <td>-0.03</td>        <td>-0.05</td>             <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.11</td>          <td>0.96</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Real personal income ex ...</td>   <td>-0.10</td>    <td>-0.05</td>        <td>-0.05</td>             <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.01</td>           <td>0.89</td>\n",
       "</tr>\n",
       "<tr>\n",
       "           <td>IP Index</td>             <td>-0.15</td>    <td>-0.09</td>        <td>0.24</td>              <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.22</td>          <td>0.20</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>IP: Final Products and N...</td>   <td>-0.15</td>    <td>-0.07</td>        <td>0.26</td>              <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.41</td>          <td>0.01</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>IP: Final Products (Mark...</td>   <td>-0.13</td>    <td>-0.07</td>        <td>0.28</td>              <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.07</td>           <td>0.04</td>\n",
       "</tr>\n",
       "<tr>\n",
       "      <td>IP: Consumer Goods</td>        <td>-0.09</td>    <td>-0.05</td>        <td>0.28</td>              <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.07</td>          <td>0.21</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>IP: Durable Consumer Goo...</td>   <td>-0.09</td>    <td>-0.07</td>        <td>0.19</td>              <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.19</td>          <td>0.53</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>IP: Nondurable Consumer ...</td>   <td>-0.05</td>    <td>-0.02</td>        <td>0.21</td>              <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.20</td>          <td>0.50</td>\n",
       "</tr>\n",
       "<tr>\n",
       "    <td>IP: Business Equipment</td>      <td>-0.13</td>    <td>-0.07</td>        <td>0.18</td>              <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.01</td>           <td>0.42</td>\n",
       "</tr>\n",
       "<tr>\n",
       "         <td>IP: Materials</td>          <td>-0.12</td>    <td>-0.10</td>        <td>0.17</td>              <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.22</td>          <td>0.56</td>\n",
       "</tr>\n",
       "<tr>\n",
       "     <td>IP: Durable Materials</td>      <td>-0.16</td>    <td>-0.07</td>        <td>0.15</td>              <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.20</td>           <td>0.34</td>\n",
       "</tr>\n",
       "<tr>\n",
       "   <td>IP: Nondurable Materials</td>     <td>-0.08</td>    <td>-0.06</td>        <td>0.14</td>              <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.14</td>          <td>0.87</td>\n",
       "</tr>\n",
       "<tr>\n",
       "    <td>IP: Manufacturing (SIC)</td>     <td>-0.16</td>    <td>-0.08</td>        <td>0.22</td>              <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.12</td>          <td>0.20</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>IP: Residential Utilitie...</td>   <td>0.01</td>     <td>-0.01</td>        <td>0.12</td>              <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.19</td>          <td>0.83</td>\n",
       "</tr>\n",
       "<tr>\n",
       "           <td>IP: Fuels</td>            <td>-0.01</td>    <td>-0.03</td>        <td>0.08</td>              <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.31</td>          <td>0.91</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Capacity Utilization: Ma...</td>   <td>-0.15</td>    <td>-0.11</td>        <td>0.22</td>              <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.04</td>          <td>0.23</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Help-Wanted Index for Un...</td>   <td>-0.06</td>    <td>-0.03</td>          <td>.</td>             <td>-0.00</td>       <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.49</td>          <td>0.72</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Ratio of Help Wanted/No....</td>   <td>-0.08</td>    <td>-0.04</td>          <td>.</td>             <td>-0.01</td>       <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.34</td>          <td>0.80</td>\n",
       "</tr>\n",
       "<tr>\n",
       "     <td>Civilian Labor Force</td>       <td>-0.03</td>    <td>0.04</td>           <td>.</td>             <td>0.03</td>        <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.17</td>          <td>0.78</td>\n",
       "</tr>\n",
       "<tr>\n",
       "      <td>Civilian Employment</td>       <td>-0.12</td>    <td>0.01</td>           <td>.</td>             <td>0.06</td>        <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.21</td>          <td>0.60</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Civilian Unemployment Ra...</td>   <td>0.13</td>     <td>0.04</td>           <td>.</td>             <td>-0.04</td>       <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.23</td>          <td>0.61</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Average Duration of Unem...</td>   <td>0.04</td>     <td>-0.03</td>          <td>.</td>             <td>-0.10</td>       <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.19</td>          <td>0.87</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Civilians Unemployed - L...</td>   <td>0.01</td>     <td>0.01</td>           <td>.</td>             <td>0.02</td>        <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.52</td>          <td>0.73</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Civilians Unemployed for...</td>   <td>0.05</td>     <td>0.03</td>           <td>.</td>             <td>0.06</td>        <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.43</td>          <td>0.76</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Civilians Unemployed - 1...</td>   <td>0.13</td>     <td>0.04</td>           <td>.</td>             <td>-0.06</td>       <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.27</td>          <td>0.64</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Civilians Unemployed for...</td>   <td>0.07</td>     <td>0.04</td>           <td>.</td>             <td>0.01</td>        <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.30</td>          <td>0.82</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Civilians Unemployed for...</td>   <td>0.10</td>     <td>0.01</td>           <td>.</td>             <td>-0.08</td>       <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.30</td>          <td>0.72</td>\n",
       "</tr>\n",
       "<tr>\n",
       "        <td>Initial Claims</td>          <td>0.07</td>     <td>0.06</td>           <td>.</td>             <td>0.08</td>        <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.32</td>          <td>0.80</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Total non...</td>   <td>-0.20</td>    <td>-0.02</td>          <td>.</td>             <td>0.16</td>        <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.12</td>          <td>0.15</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Goods-Pro...</td>   <td>-0.20</td>    <td>-0.04</td>          <td>.</td>             <td>0.17</td>        <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.01</td>           <td>0.10</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Mining an...</td>   <td>-0.08</td>    <td>-0.01</td>          <td>.</td>             <td>0.01</td>        <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.75</td>           <td>0.38</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Construct...</td>   <td>-0.17</td>    <td>0.04</td>           <td>.</td>             <td>0.15</td>        <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.07</td>           <td>0.35</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Manufactu...</td>   <td>-0.18</td>    <td>-0.09</td>          <td>.</td>             <td>0.17</td>        <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.17</td>           <td>0.11</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Durable g...</td>   <td>-0.19</td>    <td>-0.09</td>          <td>.</td>             <td>0.14</td>        <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.15</td>           <td>0.14</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Nondurabl...</td>   <td>-0.15</td>    <td>-0.10</td>          <td>.</td>             <td>0.20</td>        <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.25</td>           <td>0.29</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Service-P...</td>   <td>-0.19</td>    <td>0.00</td>           <td>.</td>             <td>0.14</td>        <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.04</td>          <td>0.30</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Trade, Tr...</td>   <td>-0.18</td>    <td>-0.03</td>          <td>.</td>             <td>0.13</td>        <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.08</td>          <td>0.32</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Wholesale...</td>   <td>-0.18</td>    <td>-0.01</td>          <td>.</td>             <td>0.12</td>        <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.24</td>           <td>0.30</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Retail Tr...</td>   <td>-0.14</td>    <td>-0.02</td>          <td>.</td>             <td>0.09</td>        <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.02</td>           <td>0.60</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Financial...</td>   <td>-0.15</td>    <td>0.04</td>           <td>.</td>             <td>0.11</td>        <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.42</td>           <td>0.43</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>All Employees: Governmen...</td>   <td>-0.01</td>    <td>0.05</td>           <td>.</td>             <td>0.00</td>        <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.09</td>          <td>0.95</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Avg Weekly Hours : Goods...</td>   <td>-0.12</td>    <td>-0.08</td>          <td>.</td>             <td>0.25</td>        <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.93</td>           <td>0.04</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Avg Weekly Overtime Hour...</td>   <td>-0.06</td>    <td>-0.06</td>          <td>.</td>             <td>-0.13</td>       <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.31</td>          <td>0.84</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Avg Weekly Hours : Manuf...</td>   <td>-0.13</td>    <td>-0.09</td>          <td>.</td>             <td>0.24</td>        <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.95</td>           <td>0.03</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Avg Hourly Earnings : Go...</td>   <td>-0.01</td>    <td>-0.01</td>          <td>.</td>             <td>-0.04</td>       <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.57</td>          <td>0.68</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Avg Hourly Earnings : Co...</td>   <td>0.00</td>     <td>-0.00</td>          <td>.</td>             <td>-0.00</td>       <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.63</td>          <td>0.61</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Avg Hourly Earnings : Ma...</td>   <td>-0.01</td>    <td>-0.02</td>          <td>.</td>             <td>-0.07</td>       <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.56</td>          <td>0.71</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Housing Starts: Total Ne...</td>   <td>-0.10</td>    <td>0.22</td>           <td>.</td>               <td>.</td>       <td>0.14</td>                     <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.00</td>           <td>0.02</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Housing Starts, Northeas...</td>   <td>-0.09</td>    <td>0.20</td>           <td>.</td>               <td>.</td>       <td>0.12</td>                     <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.16</td>           <td>0.24</td>\n",
       "</tr>\n",
       "<tr>\n",
       "    <td>Housing Starts, Midwest</td>     <td>-0.08</td>    <td>0.21</td>           <td>.</td>               <td>.</td>       <td>0.16</td>                     <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.45</td>           <td>0.12</td>\n",
       "</tr>\n",
       "<tr>\n",
       "     <td>Housing Starts, South</td>      <td>-0.11</td>    <td>0.22</td>           <td>.</td>               <td>.</td>       <td>0.14</td>                     <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.19</td>           <td>0.04</td>\n",
       "</tr>\n",
       "<tr>\n",
       "     <td>Housing Starts, West</td>       <td>-0.10</td>    <td>0.22</td>           <td>.</td>               <td>.</td>       <td>0.13</td>                     <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.12</td>           <td>0.04</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>New Private Housing Perm...</td>   <td>-0.11</td>    <td>0.22</td>           <td>.</td>               <td>.</td>       <td>0.14</td>                     <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.07</td>           <td>0.00</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>New Private Housing Perm...</td>   <td>-0.10</td>    <td>0.21</td>           <td>.</td>               <td>.</td>       <td>0.12</td>                     <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.28</td>           <td>0.14</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>New Private Housing Perm...</td>   <td>-0.09</td>    <td>0.22</td>           <td>.</td>               <td>.</td>       <td>0.16</td>                     <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.75</td>           <td>0.03</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>New Private Housing Perm...</td>   <td>-0.11</td>    <td>0.22</td>           <td>.</td>               <td>.</td>       <td>0.13</td>                     <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.45</td>           <td>0.01</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>New Private Housing Perm...</td>   <td>-0.11</td>    <td>0.22</td>           <td>.</td>               <td>.</td>       <td>0.14</td>                     <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.33</td>           <td>0.02</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Real personal consumptio...</td>   <td>-0.07</td>    <td>0.02</td>           <td>.</td>               <td>.</td>         <td>.</td>                    <td>-0.33</td>                        <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.33</td>          <td>0.62</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Real Manu. and Trade Ind...</td>   <td>-0.10</td>    <td>-0.03</td>          <td>.</td>               <td>.</td>         <td>.</td>                    <td>-0.39</td>                        <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.26</td>          <td>0.41</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Retail and Food Services...</td>   <td>-0.07</td>    <td>0.02</td>           <td>.</td>               <td>.</td>         <td>.</td>                    <td>-0.42</td>                        <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.34</td>          <td>0.45</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>New Orders for Consumer ...</td>   <td>-0.08</td>    <td>0.03</td>           <td>.</td>               <td>.</td>         <td>.</td>                    <td>-0.31</td>                        <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.12</td>           <td>0.61</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>New Orders for Durable G...</td>   <td>-0.07</td>    <td>-0.02</td>          <td>.</td>               <td>.</td>         <td>.</td>                    <td>-0.21</td>                        <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.42</td>          <td>0.72</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>New Orders for Nondefens...</td>   <td>-0.05</td>    <td>-0.02</td>          <td>.</td>               <td>.</td>         <td>.</td>                    <td>-0.15</td>                        <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.39</td>          <td>0.80</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Unfilled Orders for Dura...</td>   <td>-0.11</td>    <td>0.04</td>           <td>.</td>               <td>.</td>         <td>.</td>                    <td>0.01</td>                         <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.33</td>           <td>0.66</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Total Business Inventori...</td>   <td>-0.14</td>    <td>-0.03</td>          <td>.</td>               <td>.</td>         <td>.</td>                    <td>0.14</td>                         <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.48</td>           <td>0.40</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Total Business: Inventor...</td>   <td>0.05</td>     <td>-0.01</td>          <td>.</td>               <td>.</td>         <td>.</td>                    <td>0.50</td>                         <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.41</td>           <td>0.20</td>\n",
       "</tr>\n",
       "<tr>\n",
       "   <td>Consumer Sentiment Index</td>     <td>-0.00</td>    <td>-0.01</td>          <td>.</td>               <td>.</td>         <td>.</td>                    <td>0.04</td>                         <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.05</td>          <td>0.98</td>\n",
       "</tr>\n",
       "<tr>\n",
       "        <td>M1 Money Stock</td>          <td>0.02</td>     <td>0.02</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                        <td>-0.31</td>                   <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.61</td>          <td>0.52</td>\n",
       "</tr>\n",
       "<tr>\n",
       "        <td>M2 Money Stock</td>          <td>0.00</td>     <td>-0.00</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                        <td>-0.52</td>                   <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.40</td>          <td>0.17</td>\n",
       "</tr>\n",
       "<tr>\n",
       "      <td>Real M2 Money Stock</td>       <td>0.08</td>     <td>-0.02</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                        <td>-0.39</td>                   <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.54</td>           <td>0.42</td>\n",
       "</tr>\n",
       "<tr>\n",
       "         <td>Monetary Base</td>          <td>-0.02</td>    <td>-0.03</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                        <td>-0.26</td>                   <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.13</td>          <td>0.90</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Total Reserves of Deposi...</td>   <td>-0.01</td>    <td>-0.02</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                        <td>-0.30</td>                   <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.09</td>          <td>0.92</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Reserves Of Depository I...</td>   <td>0.07</td>     <td>-0.07</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                        <td>-0.06</td>                   <td>.</td>                 <td>.</td>         <td>.</td>               <td>0.41</td>           <td>0.77</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Commercial and Industria...</td>   <td>-0.02</td>    <td>-0.01</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                        <td>-0.12</td>                   <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.37</td>          <td>0.84</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Real Estate Loans at All...</td>   <td>0.03</td>     <td>-0.02</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                        <td>-0.18</td>                   <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.28</td>          <td>0.88</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Total Nonrevolving Credi...</td>   <td>-0.01</td>    <td>0.00</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                        <td>-0.04</td>                   <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.48</td>          <td>0.77</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Nonrevolving consumer cr...</td>   <td>0.03</td>     <td>0.01</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                        <td>0.03</td>                    <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.09</td>          <td>0.97</td>\n",
       "</tr>\n",
       "<tr>\n",
       "        <td>MZM Money Stock</td>         <td>0.01</td>     <td>-0.01</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                        <td>-0.50</td>                   <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.19</td>          <td>0.35</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Consumer Motor Vehicle L...</td>   <td>-0.00</td>    <td>-0.00</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                        <td>0.11</td>                    <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.39</td>          <td>0.83</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Total Consumer Loans and...</td>   <td>-0.01</td>    <td>-0.01</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                        <td>0.07</td>                    <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.35</td>          <td>0.88</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Securities in Bank Credi...</td>   <td>0.01</td>     <td>0.01</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                        <td>-0.11</td>                   <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.30</td>          <td>0.89</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Effective Federal Funds ...</td>   <td>-0.12</td>    <td>-0.02</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                   <td>-0.17</td>               <td>.</td>         <td>.</td>               <td>0.42</td>           <td>0.45</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>3-Month AA Financial Com...</td>   <td>-0.12</td>    <td>-0.03</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                   <td>-0.17</td>               <td>.</td>         <td>.</td>               <td>0.09</td>           <td>0.58</td>\n",
       "</tr>\n",
       "<tr>\n",
       "    <td>3-Month Treasury Bill:</td>      <td>-0.10</td>    <td>-0.04</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                   <td>-0.24</td>               <td>.</td>         <td>.</td>               <td>0.28</td>           <td>0.50</td>\n",
       "</tr>\n",
       "<tr>\n",
       "    <td>6-Month Treasury Bill:</td>      <td>-0.11</td>    <td>-0.04</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                   <td>-0.26</td>               <td>.</td>         <td>.</td>               <td>0.36</td>           <td>0.39</td>\n",
       "</tr>\n",
       "<tr>\n",
       "     <td>1-Year Treasury Rate</td>       <td>-0.10</td>    <td>-0.04</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                   <td>-0.27</td>               <td>.</td>         <td>.</td>               <td>0.30</td>           <td>0.43</td>\n",
       "</tr>\n",
       "<tr>\n",
       "     <td>5-Year Treasury Rate</td>       <td>-0.05</td>    <td>-0.01</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                   <td>-0.21</td>               <td>.</td>         <td>.</td>               <td>0.20</td>           <td>0.75</td>\n",
       "</tr>\n",
       "<tr>\n",
       "     <td>10-Year Treasury Rate</td>      <td>-0.03</td>    <td>-0.01</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                   <td>-0.17</td>               <td>.</td>         <td>.</td>               <td>0.19</td>           <td>0.84</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Moody’s Seasoned Aaa Cor...</td>   <td>-0.02</td>    <td>-0.01</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                   <td>-0.10</td>               <td>.</td>         <td>.</td>               <td>0.18</td>           <td>0.92</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Moody’s Seasoned Baa Cor...</td>   <td>-0.00</td>    <td>-0.02</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                   <td>-0.05</td>               <td>.</td>         <td>.</td>               <td>0.25</td>           <td>0.93</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>3-Month Commercial Paper...</td>   <td>0.08</td>     <td>-0.05</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                   <td>-0.12</td>               <td>.</td>         <td>.</td>               <td>0.79</td>           <td>0.27</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>3-Month Treasury C Minus...</td>   <td>-0.05</td>    <td>-0.11</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                   <td>-0.26</td>               <td>.</td>         <td>.</td>               <td>0.82</td>           <td>0.12</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>6-Month Treasury C Minus...</td>   <td>-0.04</td>    <td>-0.10</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                   <td>-0.30</td>               <td>.</td>         <td>.</td>               <td>0.88</td>           <td>0.06</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>1-Year Treasury C Minus ...</td>   <td>-0.03</td>    <td>-0.07</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                   <td>-0.32</td>               <td>.</td>         <td>.</td>               <td>0.88</td>           <td>0.04</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>5-Year Treasury C Minus ...</td>   <td>0.03</td>     <td>-0.10</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                   <td>-0.26</td>               <td>.</td>         <td>.</td>               <td>0.99</td>           <td>0.01</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>10-Year Treasury C Minus...</td>   <td>0.05</td>     <td>-0.14</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                   <td>-0.21</td>               <td>.</td>         <td>.</td>               <td>0.99</td>           <td>0.00</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Moody’s Aaa Corporate Bo...</td>   <td>0.08</td>     <td>-0.14</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                   <td>-0.17</td>               <td>.</td>         <td>.</td>               <td>1.00</td>           <td>0.00</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Moody’s Baa Corporate Bo...</td>   <td>0.12</td>     <td>-0.13</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                   <td>-0.17</td>               <td>.</td>         <td>.</td>               <td>0.99</td>           <td>0.01</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Trade Weighted U.S. Doll...</td>   <td>0.01</td>     <td>-0.05</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                   <td>-0.02</td>               <td>.</td>         <td>.</td>               <td>0.36</td>           <td>0.90</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Switzerland / U.S. Forei...</td>   <td>-0.01</td>    <td>-0.02</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                   <td>-0.05</td>               <td>.</td>         <td>.</td>               <td>0.15</td>           <td>0.99</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Japan / U.S. Foreign Exc...</td>   <td>-0.03</td>    <td>0.01</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                   <td>-0.06</td>               <td>.</td>         <td>.</td>               <td>0.23</td>           <td>0.93</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>U.S. / U.K. Foreign Exch...</td>   <td>-0.04</td>    <td>0.03</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                   <td>-0.02</td>               <td>.</td>         <td>.</td>               <td>0.25</td>           <td>0.92</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Canada / U.S. Foreign Ex...</td>   <td>0.02</td>     <td>-0.05</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                   <td>0.03</td>                <td>.</td>         <td>.</td>               <td>0.33</td>           <td>0.92</td>\n",
       "</tr>\n",
       "<tr>\n",
       "      <td>PPI: Finished Goods</td>       <td>0.00</td>     <td>0.06</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>               <td>0.26</td>        <td>.</td>               <td>-0.55</td>          <td>0.34</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>PPI: Finished Consumer G...</td>   <td>0.01</td>     <td>0.06</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>               <td>0.26</td>        <td>.</td>               <td>-0.55</td>          <td>0.33</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>PPI: Intermediate Materi...</td>   <td>0.01</td>     <td>0.07</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>               <td>0.25</td>        <td>.</td>               <td>-0.50</td>          <td>0.37</td>\n",
       "</tr>\n",
       "<tr>\n",
       "     <td>PPI: Crude Materials</td>       <td>0.01</td>     <td>0.05</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>               <td>0.17</td>        <td>.</td>               <td>-0.48</td>          <td>0.59</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Crude Oil, spliced WTI a...</td>   <td>0.02</td>     <td>0.04</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>               <td>0.14</td>        <td>.</td>               <td>-0.48</td>          <td>0.67</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>PPI: Metals and metal pr...</td>   <td>0.01</td>     <td>0.02</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>               <td>0.06</td>        <td>.</td>               <td>-0.36</td>          <td>0.85</td>\n",
       "</tr>\n",
       "<tr>\n",
       "        <td>CPI : All Items</td>         <td>0.01</td>     <td>0.08</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>               <td>0.30</td>        <td>.</td>               <td>-0.52</td>          <td>0.00</td>\n",
       "</tr>\n",
       "<tr>\n",
       "         <td>CPI : Apparel</td>          <td>-0.01</td>    <td>0.00</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>               <td>0.00</td>        <td>.</td>               <td>-0.41</td>          <td>0.83</td>\n",
       "</tr>\n",
       "<tr>\n",
       "     <td>CPI : Transportation</td>       <td>0.01</td>     <td>0.08</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>               <td>0.29</td>        <td>.</td>               <td>-0.43</td>          <td>0.06</td>\n",
       "</tr>\n",
       "<tr>\n",
       "      <td>CPI : Medical Care</td>        <td>0.01</td>     <td>-0.00</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>               <td>-0.02</td>       <td>.</td>               <td>-0.38</td>          <td>0.85</td>\n",
       "</tr>\n",
       "<tr>\n",
       "       <td>CPI : Commodities</td>        <td>0.01</td>     <td>0.08</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>               <td>0.29</td>        <td>.</td>               <td>-0.47</td>          <td>0.04</td>\n",
       "</tr>\n",
       "<tr>\n",
       "        <td>CPI : Durables</td>          <td>0.01</td>     <td>0.01</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>               <td>0.05</td>        <td>.</td>               <td>-0.23</td>          <td>0.93</td>\n",
       "</tr>\n",
       "<tr>\n",
       "        <td>CPI : Services</td>          <td>-0.00</td>    <td>0.02</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>               <td>0.06</td>        <td>.</td>               <td>-0.48</td>          <td>0.72</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>CPI : All Items Less Foo...</td>   <td>0.01</td>     <td>0.08</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>               <td>0.30</td>        <td>.</td>               <td>-0.57</td>          <td>0.01</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>CPI : All items less she...</td>   <td>0.01</td>     <td>0.08</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>               <td>0.29</td>        <td>.</td>               <td>-0.54</td>          <td>0.02</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>CPI : All items less med...</td>   <td>0.01</td>     <td>0.08</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>               <td>0.30</td>        <td>.</td>               <td>-0.69</td>          <td>0.00</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Personal Cons. Expend.: ...</td>   <td>0.01</td>     <td>0.07</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>               <td>0.26</td>        <td>.</td>               <td>-0.56</td>          <td>0.18</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Personal Cons. Exp: Dura...</td>   <td>-0.00</td>    <td>0.01</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>               <td>0.03</td>        <td>.</td>               <td>-0.40</td>          <td>0.83</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Personal Cons. Exp: Nond...</td>   <td>0.01</td>     <td>0.08</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>               <td>0.29</td>        <td>.</td>               <td>-0.49</td>          <td>0.06</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Personal Cons. Exp: Serv...</td>   <td>-0.00</td>    <td>0.01</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>               <td>0.01</td>        <td>.</td>               <td>-0.56</td>          <td>0.69</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>S&P’s Common Stock Price...</td>   <td>-0.06</td>    <td>-0.00</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>       <td>-0.48</td>             <td>-0.41</td>          <td>0.00</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>S&P’s Common Stock Price...</td>   <td>-0.06</td>    <td>-0.01</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>       <td>-0.48</td>             <td>0.29</td>           <td>0.02</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>S&P’s Composite Common S...</td>   <td>0.04</td>     <td>-0.00</td>          <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>       <td>0.48</td>              <td>0.49</td>           <td>0.09</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>S&P’s Composite Common S...</td>   <td>0.05</td>     <td>0.03</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>       <td>-0.40</td>             <td>0.86</td>           <td>0.12</td>\n",
       "</tr>\n",
       "<tr>\n",
       "              <td>VXO</td>               <td>0.17</td>     <td>0.00</td>           <td>.</td>               <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>       <td>0.21</td>              <td>0.79</td>           <td>0.11</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Real Gross Domestic Prod...</td>   <td>-0.02</td>    <td>-0.00</td>        <td>0.02</td>              <td>.</td>         <td>.</td>                      <td>.</td>                          <td>.</td>                     <td>.</td>                 <td>.</td>         <td>.</td>               <td>-0.79</td>          <td>0.06</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Transition: Factor block 0</caption>\n",
       "<tr>\n",
       "      <th></th>     <th>L1.Global.1</th> <th>L1.Global.2</th> <th>L2.Global.1</th> <th>L2.Global.2</th> <th>L3.Global.1</th> <th>L3.Global.2</th> <th>L4.Global.1</th> <th>L4.Global.2</th> <th>   error covariance</th> <th>Global.1</th> <th>Global.2</th>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Global.1</td>    <td>1.67</td>        <td>-0.88</td>       <td>-0.90</td>       <td>0.75</td>        <td>0.34</td>        <td>0.47</td>        <td>-0.18</td>       <td>-0.37</td>         <td>Global.1</td>         <td>0.74</td>     <td>0.16</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Global.2</td>    <td>0.08</td>        <td>0.94</td>        <td>-0.10</td>       <td>0.08</td>        <td>-0.03</td>       <td>0.34</td>        <td>0.01</td>        <td>-0.39</td>         <td>Global.2</td>         <td>0.16</td>     <td>0.10</td>  \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Transition: Factor block 1</caption>\n",
       "<tr>\n",
       "          <th></th>          <th>L1.Output and Income</th> <th>   error variance</th>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Output and Income</td>         <td>-0.21</td>              <td>9.49</td>       \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Transition: Factor block 2</caption>\n",
       "<tr>\n",
       "        <th></th>       <th>L1.Labor Market</th> <th>   error variance</th>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Labor Market</td>      <td>0.91</td>             <td>0.84</td>       \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Transition: Factor block 3</caption>\n",
       "<tr>\n",
       "     <th></th>     <th>L1.Housing</th> <th>   error variance</th>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Housing</td>    <td>0.98</td>          <td>0.50</td>       \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Transition: Factor block 4</caption>\n",
       "<tr>\n",
       "                    <th></th>                   <th>L1.Consumption, orders, and inventories</th> <th>   error variance</th>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Consumption, orders, and inventories</td>                  <td>-0.17</td>                        <td>2.46</td>       \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Transition: Factor block 5</caption>\n",
       "<tr>\n",
       "          <th></th>         <th>L1.Money and credit</th> <th>   error variance</th>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Money and credit</td>        <td>-0.36</td>              <td>1.99</td>       \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Transition: Factor block 6</caption>\n",
       "<tr>\n",
       "               <th></th>               <th>L1.Interest and exchange rates</th> <th>   error variance</th>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Interest and exchange rates</td>              <td>0.91</td>                    <td>0.77</td>       \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Transition: Factor block 7</caption>\n",
       "<tr>\n",
       "     <th></th>    <th>L1.Prices</th> <th>   error variance</th>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Prices</td>   <td>-0.06</td>         <td>11.76</td>      \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<caption>Transition: Factor block 8</caption>\n",
       "<tr>\n",
       "        <th></th>       <th>L1.Stock market</th> <th>   error variance</th>\n",
       "</tr>\n",
       "<tr>\n",
       "  <td>Stock market</td>      <td>0.18</td>             <td>3.74</td>       \n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Covariance matrix not calculated."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                                     Dynamic Factor Results                                     \n",
       "================================================================================================\n",
       "Dep. Variable:     \"Real Personal Income\", and 128 more   No. Observations:                  241\n",
       "Model:                             Dynamic Factor Model   Log Likelihood              -25666.313\n",
       "                               + 10 factors in 9 blocks   AIC                          52692.627\n",
       "                                + Mixed frequency (M/Q)   BIC                          55062.289\n",
       "                                  + AR(1) idiosyncratic   HQIC                         53647.320\n",
       "Date:                                  Fri, 31 Jul 2020   EM Iterations                        6\n",
       "Time:                                          01:28:24                                         \n",
       "Sample:                                      01-31-2000                                         \n",
       "                                           - 01-31-2020                                         \n",
       "Covariance Type:                           Not computed                                         \n",
       "                                                                                                          Observation equation:                                                                                                          \n",
       "=========================================================================================================================================================================================================================================\n",
       "           Factor loadings:   Global.1   Global.2 Output and Income Labor Market     Housing Consumption, orders, and inventories Money and credit Interest and exchange rates     Prices Stock market    idiosyncratic: AR(1)       var.\n",
       "-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------\n",
       "       Real Personal Income      -0.07      -0.03             -0.05            .           .                                    .                .                           .          .            .                   -0.11       0.96\n",
       "Real personal income ex ...      -0.10      -0.05             -0.05            .           .                                    .                .                           .          .            .                    0.01       0.89\n",
       "                   IP Index      -0.15      -0.09              0.24            .           .                                    .                .                           .          .            .                   -0.22       0.20\n",
       "IP: Final Products and N...      -0.15      -0.07              0.26            .           .                                    .                .                           .          .            .                   -0.41       0.01\n",
       "IP: Final Products (Mark...      -0.13      -0.07              0.28            .           .                                    .                .                           .          .            .                    0.07       0.04\n",
       "         IP: Consumer Goods      -0.09      -0.05              0.28            .           .                                    .                .                           .          .            .                   -0.07       0.21\n",
       "IP: Durable Consumer Goo...      -0.09      -0.07              0.19            .           .                                    .                .                           .          .            .                   -0.19       0.53\n",
       "IP: Nondurable Consumer ...      -0.05      -0.02              0.21            .           .                                    .                .                           .          .            .                   -0.20       0.50\n",
       "     IP: Business Equipment      -0.13      -0.07              0.18            .           .                                    .                .                           .          .            .                    0.01       0.42\n",
       "              IP: Materials      -0.12      -0.10              0.17            .           .                                    .                .                           .          .            .                   -0.22       0.56\n",
       "      IP: Durable Materials      -0.16      -0.07              0.15            .           .                                    .                .                           .          .            .                    0.20       0.34\n",
       "   IP: Nondurable Materials      -0.08      -0.06              0.14            .           .                                    .                .                           .          .            .                   -0.14       0.87\n",
       "    IP: Manufacturing (SIC)      -0.16      -0.08              0.22            .           .                                    .                .                           .          .            .                   -0.12       0.20\n",
       "IP: Residential Utilitie...       0.01      -0.01              0.12            .           .                                    .                .                           .          .            .                   -0.19       0.83\n",
       "                  IP: Fuels      -0.01      -0.03              0.08            .           .                                    .                .                           .          .            .                   -0.31       0.91\n",
       "Capacity Utilization: Ma...      -0.15      -0.11              0.22            .           .                                    .                .                           .          .            .                   -0.04       0.23\n",
       "Help-Wanted Index for Un...      -0.06      -0.03                 .        -0.00           .                                    .                .                           .          .            .                   -0.49       0.72\n",
       "Ratio of Help Wanted/No....      -0.08      -0.04                 .        -0.01           .                                    .                .                           .          .            .                   -0.34       0.80\n",
       "       Civilian Labor Force      -0.03       0.04                 .         0.03           .                                    .                .                           .          .            .                   -0.17       0.78\n",
       "        Civilian Employment      -0.12       0.01                 .         0.06           .                                    .                .                           .          .            .                   -0.21       0.60\n",
       "Civilian Unemployment Ra...       0.13       0.04                 .        -0.04           .                                    .                .                           .          .            .                   -0.23       0.61\n",
       "Average Duration of Unem...       0.04      -0.03                 .        -0.10           .                                    .                .                           .          .            .                   -0.19       0.87\n",
       "Civilians Unemployed - L...       0.01       0.01                 .         0.02           .                                    .                .                           .          .            .                   -0.52       0.73\n",
       "Civilians Unemployed for...       0.05       0.03                 .         0.06           .                                    .                .                           .          .            .                   -0.43       0.76\n",
       "Civilians Unemployed - 1...       0.13       0.04                 .        -0.06           .                                    .                .                           .          .            .                   -0.27       0.64\n",
       "Civilians Unemployed for...       0.07       0.04                 .         0.01           .                                    .                .                           .          .            .                   -0.30       0.82\n",
       "Civilians Unemployed for...       0.10       0.01                 .        -0.08           .                                    .                .                           .          .            .                   -0.30       0.72\n",
       "             Initial Claims       0.07       0.06                 .         0.08           .                                    .                .                           .          .            .                   -0.32       0.80\n",
       "All Employees: Total non...      -0.20      -0.02                 .         0.16           .                                    .                .                           .          .            .                   -0.12       0.15\n",
       "All Employees: Goods-Pro...      -0.20      -0.04                 .         0.17           .                                    .                .                           .          .            .                    0.01       0.10\n",
       "All Employees: Mining an...      -0.08      -0.01                 .         0.01           .                                    .                .                           .          .            .                    0.75       0.38\n",
       "All Employees: Construct...      -0.17       0.04                 .         0.15           .                                    .                .                           .          .            .                    0.07       0.35\n",
       "All Employees: Manufactu...      -0.18      -0.09                 .         0.17           .                                    .                .                           .          .            .                    0.17       0.11\n",
       "All Employees: Durable g...      -0.19      -0.09                 .         0.14           .                                    .                .                           .          .            .                    0.15       0.14\n",
       "All Employees: Nondurabl...      -0.15      -0.10                 .         0.20           .                                    .                .                           .          .            .                    0.25       0.29\n",
       "All Employees: Service-P...      -0.19       0.00                 .         0.14           .                                    .                .                           .          .            .                   -0.04       0.30\n",
       "All Employees: Trade, Tr...      -0.18      -0.03                 .         0.13           .                                    .                .                           .          .            .                   -0.08       0.32\n",
       "All Employees: Wholesale...      -0.18      -0.01                 .         0.12           .                                    .                .                           .          .            .                    0.24       0.30\n",
       "All Employees: Retail Tr...      -0.14      -0.02                 .         0.09           .                                    .                .                           .          .            .                    0.02       0.60\n",
       "All Employees: Financial...      -0.15       0.04                 .         0.11           .                                    .                .                           .          .            .                    0.42       0.43\n",
       "All Employees: Governmen...      -0.01       0.05                 .         0.00           .                                    .                .                           .          .            .                   -0.09       0.95\n",
       "Avg Weekly Hours : Goods...      -0.12      -0.08                 .         0.25           .                                    .                .                           .          .            .                    0.93       0.04\n",
       "Avg Weekly Overtime Hour...      -0.06      -0.06                 .        -0.13           .                                    .                .                           .          .            .                   -0.31       0.84\n",
       "Avg Weekly Hours : Manuf...      -0.13      -0.09                 .         0.24           .                                    .                .                           .          .            .                    0.95       0.03\n",
       "Avg Hourly Earnings : Go...      -0.01      -0.01                 .        -0.04           .                                    .                .                           .          .            .                   -0.57       0.68\n",
       "Avg Hourly Earnings : Co...       0.00      -0.00                 .        -0.00           .                                    .                .                           .          .            .                   -0.63       0.61\n",
       "Avg Hourly Earnings : Ma...      -0.01      -0.02                 .        -0.07           .                                    .                .                           .          .            .                   -0.56       0.71\n",
       "Housing Starts: Total Ne...      -0.10       0.22                 .            .        0.14                                    .                .                           .          .            .                    0.00       0.02\n",
       "Housing Starts, Northeas...      -0.09       0.20                 .            .        0.12                                    .                .                           .          .            .                    0.16       0.24\n",
       "    Housing Starts, Midwest      -0.08       0.21                 .            .        0.16                                    .                .                           .          .            .                    0.45       0.12\n",
       "      Housing Starts, South      -0.11       0.22                 .            .        0.14                                    .                .                           .          .            .                    0.19       0.04\n",
       "       Housing Starts, West      -0.10       0.22                 .            .        0.13                                    .                .                           .          .            .                    0.12       0.04\n",
       "New Private Housing Perm...      -0.11       0.22                 .            .        0.14                                    .                .                           .          .            .                    0.07       0.00\n",
       "New Private Housing Perm...      -0.10       0.21                 .            .        0.12                                    .                .                           .          .            .                    0.28       0.14\n",
       "New Private Housing Perm...      -0.09       0.22                 .            .        0.16                                    .                .                           .          .            .                    0.75       0.03\n",
       "New Private Housing Perm...      -0.11       0.22                 .            .        0.13                                    .                .                           .          .            .                    0.45       0.01\n",
       "New Private Housing Perm...      -0.11       0.22                 .            .        0.14                                    .                .                           .          .            .                    0.33       0.02\n",
       "Real personal consumptio...      -0.07       0.02                 .            .           .                                -0.33                .                           .          .            .                   -0.33       0.62\n",
       "Real Manu. and Trade Ind...      -0.10      -0.03                 .            .           .                                -0.39                .                           .          .            .                   -0.26       0.41\n",
       "Retail and Food Services...      -0.07       0.02                 .            .           .                                -0.42                .                           .          .            .                   -0.34       0.45\n",
       "New Orders for Consumer ...      -0.08       0.03                 .            .           .                                -0.31                .                           .          .            .                    0.12       0.61\n",
       "New Orders for Durable G...      -0.07      -0.02                 .            .           .                                -0.21                .                           .          .            .                   -0.42       0.72\n",
       "New Orders for Nondefens...      -0.05      -0.02                 .            .           .                                -0.15                .                           .          .            .                   -0.39       0.80\n",
       "Unfilled Orders for Dura...      -0.11       0.04                 .            .           .                                 0.01                .                           .          .            .                    0.33       0.66\n",
       "Total Business Inventori...      -0.14      -0.03                 .            .           .                                 0.14                .                           .          .            .                    0.48       0.40\n",
       "Total Business: Inventor...       0.05      -0.01                 .            .           .                                 0.50                .                           .          .            .                    0.41       0.20\n",
       "   Consumer Sentiment Index      -0.00      -0.01                 .            .           .                                 0.04                .                           .          .            .                   -0.05       0.98\n",
       "             M1 Money Stock       0.02       0.02                 .            .           .                                    .            -0.31                           .          .            .                   -0.61       0.52\n",
       "             M2 Money Stock       0.00      -0.00                 .            .           .                                    .            -0.52                           .          .            .                   -0.40       0.17\n",
       "        Real M2 Money Stock       0.08      -0.02                 .            .           .                                    .            -0.39                           .          .            .                    0.54       0.42\n",
       "              Monetary Base      -0.02      -0.03                 .            .           .                                    .            -0.26                           .          .            .                   -0.13       0.90\n",
       "Total Reserves of Deposi...      -0.01      -0.02                 .            .           .                                    .            -0.30                           .          .            .                   -0.09       0.92\n",
       "Reserves Of Depository I...       0.07      -0.07                 .            .           .                                    .            -0.06                           .          .            .                    0.41       0.77\n",
       "Commercial and Industria...      -0.02      -0.01                 .            .           .                                    .            -0.12                           .          .            .                   -0.37       0.84\n",
       "Real Estate Loans at All...       0.03      -0.02                 .            .           .                                    .            -0.18                           .          .            .                   -0.28       0.88\n",
       "Total Nonrevolving Credi...      -0.01       0.00                 .            .           .                                    .            -0.04                           .          .            .                   -0.48       0.77\n",
       "Nonrevolving consumer cr...       0.03       0.01                 .            .           .                                    .             0.03                           .          .            .                   -0.09       0.97\n",
       "            MZM Money Stock       0.01      -0.01                 .            .           .                                    .            -0.50                           .          .            .                   -0.19       0.35\n",
       "Consumer Motor Vehicle L...      -0.00      -0.00                 .            .           .                                    .             0.11                           .          .            .                   -0.39       0.83\n",
       "Total Consumer Loans and...      -0.01      -0.01                 .            .           .                                    .             0.07                           .          .            .                   -0.35       0.88\n",
       "Securities in Bank Credi...       0.01       0.01                 .            .           .                                    .            -0.11                           .          .            .                   -0.30       0.89\n",
       "Effective Federal Funds ...      -0.12      -0.02                 .            .           .                                    .                .                       -0.17          .            .                    0.42       0.45\n",
       "3-Month AA Financial Com...      -0.12      -0.03                 .            .           .                                    .                .                       -0.17          .            .                    0.09       0.58\n",
       "     3-Month Treasury Bill:      -0.10      -0.04                 .            .           .                                    .                .                       -0.24          .            .                    0.28       0.50\n",
       "     6-Month Treasury Bill:      -0.11      -0.04                 .            .           .                                    .                .                       -0.26          .            .                    0.36       0.39\n",
       "       1-Year Treasury Rate      -0.10      -0.04                 .            .           .                                    .                .                       -0.27          .            .                    0.30       0.43\n",
       "       5-Year Treasury Rate      -0.05      -0.01                 .            .           .                                    .                .                       -0.21          .            .                    0.20       0.75\n",
       "      10-Year Treasury Rate      -0.03      -0.01                 .            .           .                                    .                .                       -0.17          .            .                    0.19       0.84\n",
       "Moody’s Seasoned Aaa Cor...      -0.02      -0.01                 .            .           .                                    .                .                       -0.10          .            .                    0.18       0.92\n",
       "Moody’s Seasoned Baa Cor...      -0.00      -0.02                 .            .           .                                    .                .                       -0.05          .            .                    0.25       0.93\n",
       "3-Month Commercial Paper...       0.08      -0.05                 .            .           .                                    .                .                       -0.12          .            .                    0.79       0.27\n",
       "3-Month Treasury C Minus...      -0.05      -0.11                 .            .           .                                    .                .                       -0.26          .            .                    0.82       0.12\n",
       "6-Month Treasury C Minus...      -0.04      -0.10                 .            .           .                                    .                .                       -0.30          .            .                    0.88       0.06\n",
       "1-Year Treasury C Minus ...      -0.03      -0.07                 .            .           .                                    .                .                       -0.32          .            .                    0.88       0.04\n",
       "5-Year Treasury C Minus ...       0.03      -0.10                 .            .           .                                    .                .                       -0.26          .            .                    0.99       0.01\n",
       "10-Year Treasury C Minus...       0.05      -0.14                 .            .           .                                    .                .                       -0.21          .            .                    0.99       0.00\n",
       "Moody’s Aaa Corporate Bo...       0.08      -0.14                 .            .           .                                    .                .                       -0.17          .            .                    1.00       0.00\n",
       "Moody’s Baa Corporate Bo...       0.12      -0.13                 .            .           .                                    .                .                       -0.17          .            .                    0.99       0.01\n",
       "Trade Weighted U.S. Doll...       0.01      -0.05                 .            .           .                                    .                .                       -0.02          .            .                    0.36       0.90\n",
       "Switzerland / U.S. Forei...      -0.01      -0.02                 .            .           .                                    .                .                       -0.05          .            .                    0.15       0.99\n",
       "Japan / U.S. Foreign Exc...      -0.03       0.01                 .            .           .                                    .                .                       -0.06          .            .                    0.23       0.93\n",
       "U.S. / U.K. Foreign Exch...      -0.04       0.03                 .            .           .                                    .                .                       -0.02          .            .                    0.25       0.92\n",
       "Canada / U.S. Foreign Ex...       0.02      -0.05                 .            .           .                                    .                .                        0.03          .            .                    0.33       0.92\n",
       "        PPI: Finished Goods       0.00       0.06                 .            .           .                                    .                .                           .       0.26            .                   -0.55       0.34\n",
       "PPI: Finished Consumer G...       0.01       0.06                 .            .           .                                    .                .                           .       0.26            .                   -0.55       0.33\n",
       "PPI: Intermediate Materi...       0.01       0.07                 .            .           .                                    .                .                           .       0.25            .                   -0.50       0.37\n",
       "       PPI: Crude Materials       0.01       0.05                 .            .           .                                    .                .                           .       0.17            .                   -0.48       0.59\n",
       "Crude Oil, spliced WTI a...       0.02       0.04                 .            .           .                                    .                .                           .       0.14            .                   -0.48       0.67\n",
       "PPI: Metals and metal pr...       0.01       0.02                 .            .           .                                    .                .                           .       0.06            .                   -0.36       0.85\n",
       "            CPI : All Items       0.01       0.08                 .            .           .                                    .                .                           .       0.30            .                   -0.52       0.00\n",
       "              CPI : Apparel      -0.01       0.00                 .            .           .                                    .                .                           .       0.00            .                   -0.41       0.83\n",
       "       CPI : Transportation       0.01       0.08                 .            .           .                                    .                .                           .       0.29            .                   -0.43       0.06\n",
       "         CPI : Medical Care       0.01      -0.00                 .            .           .                                    .                .                           .      -0.02            .                   -0.38       0.85\n",
       "          CPI : Commodities       0.01       0.08                 .            .           .                                    .                .                           .       0.29            .                   -0.47       0.04\n",
       "             CPI : Durables       0.01       0.01                 .            .           .                                    .                .                           .       0.05            .                   -0.23       0.93\n",
       "             CPI : Services      -0.00       0.02                 .            .           .                                    .                .                           .       0.06            .                   -0.48       0.72\n",
       "CPI : All Items Less Foo...       0.01       0.08                 .            .           .                                    .                .                           .       0.30            .                   -0.57       0.01\n",
       "CPI : All items less she...       0.01       0.08                 .            .           .                                    .                .                           .       0.29            .                   -0.54       0.02\n",
       "CPI : All items less med...       0.01       0.08                 .            .           .                                    .                .                           .       0.30            .                   -0.69       0.00\n",
       "Personal Cons. Expend.: ...       0.01       0.07                 .            .           .                                    .                .                           .       0.26            .                   -0.56       0.18\n",
       "Personal Cons. Exp: Dura...      -0.00       0.01                 .            .           .                                    .                .                           .       0.03            .                   -0.40       0.83\n",
       "Personal Cons. Exp: Nond...       0.01       0.08                 .            .           .                                    .                .                           .       0.29            .                   -0.49       0.06\n",
       "Personal Cons. Exp: Serv...      -0.00       0.01                 .            .           .                                    .                .                           .       0.01            .                   -0.56       0.69\n",
       "S&P’s Common Stock Price...      -0.06      -0.00                 .            .           .                                    .                .                           .          .        -0.48                   -0.41       0.00\n",
       "S&P’s Common Stock Price...      -0.06      -0.01                 .            .           .                                    .                .                           .          .        -0.48                    0.29       0.02\n",
       "S&P’s Composite Common S...       0.04      -0.00                 .            .           .                                    .                .                           .          .         0.48                    0.49       0.09\n",
       "S&P’s Composite Common S...       0.05       0.03                 .            .           .                                    .                .                           .          .        -0.40                    0.86       0.12\n",
       "                        VXO       0.17       0.00                 .            .           .                                    .                .                           .          .         0.21                    0.79       0.11\n",
       "Real Gross Domestic Prod...      -0.02      -0.00              0.02            .           .                                    .                .                           .          .            .                   -0.79       0.06\n",
       "                                                              Transition: Factor block 0                                                              \n",
       "======================================================================================================================================================\n",
       "           L1.Global.1 L1.Global.2 L2.Global.1 L2.Global.2 L3.Global.1 L3.Global.2 L4.Global.1 L4.Global.2    error covariance    Global.1    Global.2\n",
       "------------------------------------------------------------------------------------------------------------------------------------------------------\n",
       "  Global.1        1.67       -0.88       -0.90        0.75        0.34        0.47       -0.18       -0.37            Global.1        0.74        0.16\n",
       "  Global.2        0.08        0.94       -0.10        0.08       -0.03        0.34        0.01       -0.39            Global.2        0.16        0.10\n",
       "               Transition: Factor block 1               \n",
       "========================================================\n",
       "                  L1.Output and Income    error variance\n",
       "--------------------------------------------------------\n",
       "Output and Income                -0.21              9.49\n",
       "          Transition: Factor block 2          \n",
       "==============================================\n",
       "             L1.Labor Market    error variance\n",
       "----------------------------------------------\n",
       "Labor Market            0.91              0.84\n",
       "       Transition: Factor block 3      \n",
       "=======================================\n",
       "           L1.Housing    error variance\n",
       "---------------------------------------\n",
       "   Housing       0.98              0.50\n",
       "                                  Transition: Factor block 4                                  \n",
       "==============================================================================================\n",
       "                                     L1.Consumption, orders, and inventories    error variance\n",
       "----------------------------------------------------------------------------------------------\n",
       "Consumption, orders, and inventories                                   -0.17              2.46\n",
       "              Transition: Factor block 5              \n",
       "======================================================\n",
       "                 L1.Money and credit    error variance\n",
       "------------------------------------------------------\n",
       "Money and credit               -0.36              1.99\n",
       "                         Transition: Factor block 6                         \n",
       "============================================================================\n",
       "                            L1.Interest and exchange rates    error variance\n",
       "----------------------------------------------------------------------------\n",
       "Interest and exchange rates                           0.91              0.77\n",
       "       Transition: Factor block 7      \n",
       "=======================================\n",
       "            L1.Prices    error variance\n",
       "---------------------------------------\n",
       "    Prices      -0.06             11.76\n",
       "          Transition: Factor block 8          \n",
       "==============================================\n",
       "             L1.Stock market    error variance\n",
       "----------------------------------------------\n",
       "Stock market            0.18              3.74\n",
       "==============================================\n",
       "\n",
       "Warnings:\n",
       "[1] Covariance matrix not calculated.\n",
       "\"\"\""
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Estimated factors\n",
    "\n",
    "In addition to the estimates of the parameters, the `results` object contains the estimates of the latent factors. These are most conveniently accessed through the `factors` attribute. This attribute in turn contains four sub-attributes:\n",
    "\n",
    "- `smoothed`: estimates of the factors, conditional on the full dataset (also called \"smoothed\" or \"two-sided\" estimates)\n",
    "- `smoothed_cov`: covariance matrix of the factor estimates, conditional on the full dataset\n",
    "- `filtered`: estimates of the factors, where the estimate at time $t$ only uses information through time $t$ (also called \"filtered\" or \"one-sided\" estimates\n",
    "- `filtered_cov`: covariance matrix of the factor estimates, where the estimate at time $t$ only uses information through time $t$\n",
    "\n",
    "As an example, in the next cell we plot three of the smoothed factors and 95% confidence intervals.\n",
    "\n",
    "**Note**: The estimated factors are not identified without additional assumptions that this model does not impose (see for example Bańbura and Modugno, 2014, for details). As a result, it can be difficult to interpet the factors themselves. (Despite this, the space spanned by the factors *is* identified, so that forecasting and nowcasting exercises, like those we discuss later, are unambiguous).\n",
    "\n",
    "For example, in the plot\n",
    "below, the \"Global.1\" factor increases markedly in 2009, following the global financial crisis. However, many of the factor loadings in the summary above are negative – for example, this is true of the output, consumption, and income series. Therefore the increase in the \"Global.1\" factor during this period actually implies a strong *decrease* in output, consumption, and income."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Get estimates of the global and labor market factors,\n",
    "# conditional on the full dataset (\"smoothed\")\n",
    "factor_names = ['Global.1', 'Global.2', 'Labor Market']\n",
    "mean = results.factors.smoothed[factor_names]\n",
    "\n",
    "# Compute 95% confidence intervals\n",
    "from scipy.stats import norm\n",
    "std = pd.concat([results.factors.smoothed_cov.loc[name, name]\n",
    "                 for name in factor_names], axis=1)\n",
    "crit = norm.ppf(1 - 0.05 / 2)\n",
    "lower = mean - crit * std\n",
    "upper = mean + crit * std\n",
    "\n",
    "with sns.color_palette('deep'):\n",
    "    fig, ax = plt.subplots(figsize=(14, 3))\n",
    "    mean.plot(ax=ax)\n",
    "    \n",
    "    for name in factor_names:\n",
    "        ax.fill_between(mean.index, lower[name], upper[name], alpha=0.3)\n",
    "    \n",
    "    ax.set(title='Estimated factors: smoothed estimates and 95% confidence intervals')\n",
    "    fig.tight_layout();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Explanatory power of the factors\n",
    "\n",
    "One way to examine how the factors relate to the observed variables is to compute the explanatory power that each factor has for each variable, by regressing each variable on a constant plus one or more of the smoothed factor estimates and storing the resulting $R^2$, or \"coefficient of determination\", value.\n",
    "\n",
    "**Computing $R^2$**\n",
    "\n",
    "The `get_coefficients_of_determination` method in the results object has three options for the `method` argument:\n",
    "\n",
    "- `method='individual'` retrieves the $R^2$ value for each observed variable regressed on each individual factor (plus a constant term)\n",
    "- `method='joint'` retrieves the $R^2$ value for each observed variable regressed on all factors that the variable loads on\n",
    "- `method='cumulative'` retrieves the $R^2$ value for each observed variable regressed on an expanding set of factors. The expanding set begins with the $R^2$ from a regression of each variable on the first factor that the variable loads on (as it appears in, for example, the summary tables above) plus a constant. For the next factor in the list, the $R^2$ is computed by a regression on the first two factors (assuming that a given variable loads on both factors).\n",
    "\n",
    "**Example:** top 10 variables explained by the global factors\n",
    "\n",
    "Below, we compute according to the `method='individual'` approach, and then show the top 10 observed variables that are explained (individually) by each of the two global factors.\n",
    "\n",
    "- The first global factor explains labor market series well, but also includes Real GDP and a measure of stock market volatility (VXO)\n",
    "- The second factor appears to largely explain housing-related variables (in fact, this might be an argument for dropping the \"Housing\" group-specific factor)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"2\" halign=\"left\">Top ten variables explained by Global.1</th>\n",
       "      <th colspan=\"2\" halign=\"left\">Top ten variables explained by Global.2</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>Variable</th>\n",
       "      <th>$R^2$</th>\n",
       "      <th>Variable</th>\n",
       "      <th>$R^2$</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>All Employees: Total nonfarm</td>\n",
       "      <td>0.74</td>\n",
       "      <td>Housing Starts: Total New Privately Owned</td>\n",
       "      <td>0.65</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>All Employees: Goods-Producing Industries</td>\n",
       "      <td>0.73</td>\n",
       "      <td>New Private Housing Permits, South (SAAR)</td>\n",
       "      <td>0.65</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>All Employees: Durable goods</td>\n",
       "      <td>0.66</td>\n",
       "      <td>Housing Starts, South</td>\n",
       "      <td>0.65</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>All Employees: Manufacturing</td>\n",
       "      <td>0.65</td>\n",
       "      <td>New Private Housing Permits (SAAR)</td>\n",
       "      <td>0.65</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>All Employees: Wholesale Trade</td>\n",
       "      <td>0.64</td>\n",
       "      <td>New Private Housing Permits, West (SAAR)</td>\n",
       "      <td>0.64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>All Employees: Service-Providing Industries</td>\n",
       "      <td>0.63</td>\n",
       "      <td>Housing Starts, West</td>\n",
       "      <td>0.64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>All Employees: Trade, Transportation &amp; Utilities</td>\n",
       "      <td>0.61</td>\n",
       "      <td>New Private Housing Permits, Midwest (SAAR)</td>\n",
       "      <td>0.56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>VXO</td>\n",
       "      <td>0.61</td>\n",
       "      <td>New Private Housing Permits, Northeast (SAAR)</td>\n",
       "      <td>0.54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Real Gross Domestic Product, 3 Decimal (Billio...</td>\n",
       "      <td>0.53</td>\n",
       "      <td>Housing Starts, Midwest</td>\n",
       "      <td>0.52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>All Employees: Construction</td>\n",
       "      <td>0.52</td>\n",
       "      <td>Housing Starts, Northeast</td>\n",
       "      <td>0.52</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Top ten variables explained by Global.1        \\\n",
       "                                            Variable $R^2$   \n",
       "0                       All Employees: Total nonfarm  0.74   \n",
       "1          All Employees: Goods-Producing Industries  0.73   \n",
       "2                       All Employees: Durable goods  0.66   \n",
       "3                       All Employees: Manufacturing  0.65   \n",
       "4                     All Employees: Wholesale Trade  0.64   \n",
       "5        All Employees: Service-Providing Industries  0.63   \n",
       "6   All Employees: Trade, Transportation & Utilities  0.61   \n",
       "7                                                VXO  0.61   \n",
       "8  Real Gross Domestic Product, 3 Decimal (Billio...  0.53   \n",
       "9                        All Employees: Construction  0.52   \n",
       "\n",
       "         Top ten variables explained by Global.2        \n",
       "                                        Variable $R^2$  \n",
       "0      Housing Starts: Total New Privately Owned  0.65  \n",
       "1      New Private Housing Permits, South (SAAR)  0.65  \n",
       "2                          Housing Starts, South  0.65  \n",
       "3             New Private Housing Permits (SAAR)  0.65  \n",
       "4       New Private Housing Permits, West (SAAR)  0.64  \n",
       "5                           Housing Starts, West  0.64  \n",
       "6    New Private Housing Permits, Midwest (SAAR)  0.56  \n",
       "7  New Private Housing Permits, Northeast (SAAR)  0.54  \n",
       "8                        Housing Starts, Midwest  0.52  \n",
       "9                      Housing Starts, Northeast  0.52  "
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rsquared = results.get_coefficients_of_determination(method='individual')\n",
    "\n",
    "top_ten = []\n",
    "for factor_name in rsquared.columns[:2]:\n",
    "    top_factor = (rsquared[factor_name].sort_values(ascending=False)\n",
    "                                       .iloc[:10].round(2).reset_index())\n",
    "    top_factor.columns = pd.MultiIndex.from_product([\n",
    "        [f'Top ten variables explained by {factor_name}'],\n",
    "        ['Variable', r'$R^2$']])\n",
    "    top_ten.append(top_factor)\n",
    "pd.concat(top_ten, axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Plotting $R^2$**\n",
    "\n",
    "When there are a large number of observed variables, it is often easier to plot the $R^2$ values for each variable. This can be done using the `plot_coefficients_of_determination` method in the results object. It accepts the same `method` arguments as the `get_coefficients_of_determination` method, above.\n",
    "\n",
    "Below, we plot the $R^2$ values from the \"individual\" regressions, for each factor. Because there are so many variables, this graphical tool is best for identifying trends overall and within groups, and we do not display the names of the variables on the x-axis label."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x648 with 10 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "with sns.color_palette('deep'):\n",
    "    fig = results.plot_coefficients_of_determination(method='individual', figsize=(14, 9))\n",
    "    fig.suptitle(r'$R^2$ - regression on individual factors', fontsize=14, fontweight=600)\n",
    "    fig.tight_layout(rect=[0, 0, 1, 0.95]);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Alternatively, we might look at the overall explanatory value to a given variable of all factors that the variable loads on. To do that, we can use the same function but with `method='joint'` .\n",
    "\n",
    "To make it easier to identify patterns, we add in shading and labels to identify the different groups of variables, as well as our only quarterly variable, GDP."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x216 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "group_counts = defn_m[['description', 'group']]\n",
    "group_counts = group_counts[group_counts['description'].isin(dta['2020-02'].dta_m.columns)]\n",
    "group_counts = group_counts.groupby('group', sort=False).count()['description'].cumsum()\n",
    "\n",
    "with sns.color_palette('deep'):\n",
    "    fig = results.plot_coefficients_of_determination(method='joint', figsize=(14, 3));\n",
    "\n",
    "    # Add in group labels\n",
    "    ax = fig.axes[0]\n",
    "    ax.set_ylim(0, 1.2)\n",
    "    for i in np.arange(1, len(group_counts), 2):\n",
    "        start = 0 if i == 0 else group_counts[i - 1]\n",
    "        end = group_counts[i] + 1\n",
    "        ax.fill_between(np.arange(start, end) - 0.6, 0, 1.2, color='k', alpha=0.1)\n",
    "    for i in range(len(group_counts)):\n",
    "        start = 0 if i == 0 else group_counts[i - 1]\n",
    "        end = group_counts[i]\n",
    "        n = end - start\n",
    "        text = group_counts.index[i]\n",
    "        if len(text) > n:\n",
    "            text = text[:n - 3] + '...'\n",
    "\n",
    "        ax.annotate(text, (start + n / 2, 1.1), ha='center')\n",
    "\n",
    "    # Add label for GDP\n",
    "    ax.set_xlim(-1.5, model.k_endog + 0.5)\n",
    "    ax.annotate('GDP', (model.k_endog - 1.1, 1.05), ha='left', rotation=90)\n",
    "\n",
    "    fig.tight_layout();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Forecasting\n",
    "\n",
    "One of the benefits of these models is that we can use the dynamics of the factors to produce forecasts of any of the observed variables. This is straightforward here, using the `forecast` or `get_forecast` results methods. These take a single argument, which must be either:\n",
    "\n",
    "- an integer, specifying the number of steps ahead to forecast\n",
    "- a date, specifying the date of the final forecast to make\n",
    "\n",
    "The `forecast` method only produces a series of point forecasts for all of the observed variables, while the `get_forecast` method returns a new forecast results object, that can also be used to compute confidence intervals. \n",
    "\n",
    "**Note**: these forecasts are in the same scale as the variables passed to the `DynamicFactorMQ` constructor, even if `standardize=True` has been used.\n",
    "\n",
    "Below is an example of the `forecast` method."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                                2020-02   2020-03   2020-04\n",
      "Real Personal Income                           0.220708  0.271456  0.257587\n",
      "Real personal income ex transfer receipts      0.232887  0.270109  0.255253\n",
      "IP Index                                       0.251309  0.156593  0.161717\n",
      "IP: Final Products and Nonindustrial Supplies  0.274381  0.113899  0.134390\n",
      "IP: Final Products (Market Group)              0.279685  0.097909  0.128986\n"
     ]
    }
   ],
   "source": [
    "# Create point forecasts, 3 steps ahead\n",
    "point_forecasts = results.forecast(steps=3)\n",
    "\n",
    "# Print the forecasts for the first 5 observed variables\n",
    "print(point_forecasts.T.head())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In addition to `forecast` and `get_forecast`, there are two more general methods, `predict` and `get_prediction` that allow for both of in-sample prediction and out-of-sample forecasting. Instead of a `steps` argument, they take `start` and `end` arguments, which can be either in-sample dates or out-of-sample dates.\n",
    "\n",
    "Below, we give an example of using `get_prediction` to show in-sample predictions and out-of-sample forecasts for some spreads between Treasury securities and the Federal Funds Rate."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Create forecasts results objects, through the end of 20201\n",
    "prediction_results = results.get_prediction(start='2000', end='2022')\n",
    "\n",
    "variables = ['1-Year Treasury C Minus FEDFUNDS',\n",
    "             '5-Year Treasury C Minus FEDFUNDS',\n",
    "             '10-Year Treasury C Minus FEDFUNDS']\n",
    "\n",
    "# The `predicted_mean` attribute gives the same\n",
    "# point forecasts that would have been returned from\n",
    "# using the `predict` or `forecast` methods.\n",
    "point_predictions = prediction_results.predicted_mean[variables]\n",
    "\n",
    "# We can use the `conf_int` method to get confidence\n",
    "# intervals; here, the 95% confidence interval\n",
    "ci = prediction_results.conf_int(alpha=0.05)\n",
    "lower = ci[[f'lower {name}' for name in variables]]\n",
    "upper = ci[[f'upper {name}' for name in variables]]\n",
    "\n",
    "# Plot the forecasts and confidence intervals\n",
    "with sns.color_palette('deep'):\n",
    "    fig, ax = plt.subplots(figsize=(14, 4))\n",
    "\n",
    "    # Plot the in-sample predictions\n",
    "    point_predictions.loc[:'2020-01'].plot(ax=ax)\n",
    "\n",
    "    # Plot the out-of-sample forecasts\n",
    "    point_predictions.loc['2020-01':].plot(ax=ax, linestyle='--',\n",
    "                                           color=['C0', 'C1', 'C2'],\n",
    "                                           legend=False)\n",
    "\n",
    "    # Confidence intervals\n",
    "    for name in variables:\n",
    "        ax.fill_between(ci.index,\n",
    "                        lower[f'lower {name}'],\n",
    "                        upper[f'upper {name}'], alpha=0.1)\n",
    "        \n",
    "    # Forecast period, set title\n",
    "    ylim = ax.get_ylim()\n",
    "    ax.vlines('2020-01', ylim[0], ylim[1], linewidth=1)\n",
    "    ax.annotate(r' Forecast $\\rightarrow$', ('2020-01', -1.7))\n",
    "    ax.set(title=('Treasury securities / Federal Funds Rate spreads:'\n",
    "                  ' in-sample predictions and out-of-sample forecasts, with 95% confidence intervals'), ylim=ylim)\n",
    "    \n",
    "    fig.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Forecasting example\n",
    "\n",
    "The variables that we showed in the forecasts above were not transformed from their original values. As a result, the predictions were already interpretable as spreads. For the other observed variables that were transformed prior to construting the model, our forecasts will be in the transformed scale.\n",
    "\n",
    "For example, although the original data in the FRED-MD/QD datasets for Real GDP is in \"Billions of Chained 2012 Dollars\", this variable was transformed to the annualized quarterly growth rate (percent change) for inclusion in the model. Similarly, the Civilian Unemployment Rate was originally in \"Percent\", but it was transformed into the 1-month change (first difference) for inclusion in the model.\n",
    "\n",
    "Because the transformed data was provided to the model, the prediction and forecasting methods will produce predictions and forecasts in the transformed space. (Reminder: the transformation step, which we did prior to constructing the model, is different from the standardization step, which the model handles automatically, and which we do not need to manually reverse).\n",
    "\n",
    "Below, we compute and plot the forecasts directly from the model associated with real GDP and the unemployment rate."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get the titles of the variables as they appear in the dataset\n",
    "unemp_description = 'Civilian Unemployment Rate'\n",
    "gdp_description = 'Real Gross Domestic Product, 3 Decimal (Billions of Chained 2012 Dollars)'\n",
    "\n",
    "# Compute the point forecasts\n",
    "fcast_m = results.forecast('2021-12')[unemp_description]\n",
    "fcast_q = results.forecast('2021-12')[gdp_description].resample('Q').last()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# For more convenient plotting, combine the observed data with the forecasts\n",
    "plot_m = pd.concat([dta['2020-02'].dta_m.loc['2000':, unemp_description], fcast_m])\n",
    "plot_q = pd.concat([dta['2020-02'].dta_q.loc['2000':, gdp_description], fcast_q])\n",
    "\n",
    "with sns.color_palette('deep'):\n",
    "    fig, axes = plt.subplots(2, figsize=(14, 4))\n",
    "\n",
    "    # Plot real GDP growth, data and forecasts\n",
    "    plot_q.plot(ax=axes[0])\n",
    "    axes[0].set(title='Real Gross Domestic Product (transformed: annualized growth rate)')\n",
    "    axes[0].hlines(0, plot_q.index[0], plot_q.index[-1], linewidth=1)\n",
    "\n",
    "    # Plot the change in the unemployment rate, data and forecasts\n",
    "    plot_m.plot(ax=axes[1])\n",
    "    axes[1].set(title='Civilian Unemployment Rate (transformed: change)')\n",
    "    axes[1].hlines(0, plot_m.index[0], plot_m.index[-1], linewidth=1)\n",
    "    \n",
    "    # Show the forecast period in each graph\n",
    "    for i in range(2):\n",
    "        ylim = axes[i].get_ylim()\n",
    "        axes[i].fill_between(plot_q.loc['2020-02':].index,\n",
    "                             ylim[0], ylim[1], alpha=0.1, color='C0')\n",
    "        axes[i].annotate(r' Forecast $\\rightarrow$',\n",
    "                         ('2020-03', ylim[0] + 0.1 * ylim[1]))\n",
    "        axes[i].set_ylim(ylim)\n",
    "\n",
    "    # Title\n",
    "    fig.suptitle('Data and forecasts (February 2020 vintage), transformed scale',\n",
    "                 fontsize=14, fontweight=600)\n",
    "\n",
    "    fig.tight_layout(rect=[0, 0, 1, 0.95]);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For point forecasts, we can also reverse the transformations to get point forecasts in the original scale.\n",
    "\n",
    "**Aside**: for non-linear transformations, it would **not** be valid to compute confidence intervals in the original space by reversing the transformation on the confidence intervals computed for the transformed space."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Reverse the transformations\n",
    "\n",
    "# For real GDP, we take the level in 2000Q1 from the original data,\n",
    "# and then apply the growth rates to compute the remaining levels\n",
    "plot_q_orig = (plot_q / 100 + 1)**0.25\n",
    "plot_q_orig.loc['2000Q1'] = dta['2020-02'].orig_q.loc['2000Q1', gdp_description]\n",
    "plot_q_orig = plot_q_orig.cumprod()\n",
    "\n",
    "# For the unemployment rate, we take the level in 2000-01 from\n",
    "# the original data, and then we apply the changes to compute the\n",
    "# remaining levels\n",
    "plot_m_orig = plot_m.copy()\n",
    "plot_m_orig.loc['2000-01'] = dta['2020-02'].orig_m.loc['2000-01', unemp_description]\n",
    "plot_m_orig = plot_m_orig.cumsum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "with sns.color_palette('deep'):\n",
    "    fig, axes = plt.subplots(2, figsize=(14, 4))\n",
    "\n",
    "    # Plot real GDP, data and forecasts\n",
    "    plot_q_orig.plot(ax=axes[0])\n",
    "    axes[0].set(title=('Real Gross Domestic Product'\n",
    "                       ' (original scale: Billions of Chained 2012 Dollars)'))\n",
    "\n",
    "    # Plot the unemployment rate, data and forecasts\n",
    "    plot_m_orig.plot(ax=axes[1])\n",
    "    axes[1].set(title='Civilian Unemployment Rate (original scale: Percent)')\n",
    "\n",
    "    # Show the forecast period in each graph\n",
    "    for i in range(2):\n",
    "        ylim = axes[i].get_ylim()\n",
    "        axes[i].fill_between(plot_q.loc['2020-02':].index,\n",
    "                             ylim[0], ylim[1], alpha=0.1, color='C0')\n",
    "        axes[i].annotate(r' Forecast $\\rightarrow$',\n",
    "                         ('2020-03', ylim[0] + 0.5 * (ylim[1] - ylim[0])))\n",
    "        axes[i].set_ylim(ylim)\n",
    "\n",
    "    # Title\n",
    "    fig.suptitle('Data and forecasts (February 2020 vintage), original scale',\n",
    "                 fontsize=14, fontweight=600)\n",
    "\n",
    "    fig.tight_layout(rect=[0, 0, 1, 0.95]);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Nowcasting GDP, real-time forecast updates, and the news\n",
    "\n",
    "The forecasting exercises above were based on our baseline results object, which was computed using the February 2020 vintage of data. This was prior to any effect from the COVID-19 pandemic, and near the end of a historically long economic expansion. As a result, the forecasts above paint a relatively rosy picture for the economy, with strong real GDP growth and a continued decline in the unemployment rate. However, the economic data for March through June (which is the last vintage that was available at the time this notebook was produced) showed strong negative economic effects stemming from the pandemic and the associated disruptions to economic activity.\n",
    "\n",
    "It is straightforward to update our model to take into account new data, and to produce new forecasts. Moreover, we can compute the effect that each new observation has on our forecasts. To illustrate, we consider the exercise of forecasting real GDP growth in 2020Q2. Since this is the current quarter for most of this period, this is an example of \"nowcasting\"."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Baseline GDP forecast: February 2020 vintage**\n",
    "\n",
    "To begin with, we examine the forecast of our model for real GDP growth in 2020Q2. This model is a mixed frequency model that is estimated at the monthly frequency, and the estimates for quarterly variables correspond to the last months of each quarter. As a result, we're interested in the forecast for June 2020."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Baseline (February 2020) forecast for real GDP growth in 2020Q2: 2.68%\n"
     ]
    }
   ],
   "source": [
    "# The original point forecasts are monthly\n",
    "point_forecasts_m = results.forecast('June 2020')[gdp_description]\n",
    "\n",
    "# Resample to quarterly frequency by taking the value in the last\n",
    "# month of each quarter\n",
    "point_forecasts_q = point_forecasts_m.resample('Q').last()\n",
    "\n",
    "print('Baseline (February 2020) forecast for real GDP growth'\n",
    "      f' in 2020Q2: {point_forecasts_q[\"2020Q2\"]:.2f}%')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Updated GDP forecast: March 2020 vintage**\n",
    "\n",
    "Next, we consider taking into account data for the next available vintage, which is March 2020. Note that for the March 2020 vintage, the reference month of the released data still only covers the period through February 2020. This is still before the economic effects of the pandemic, so we expect to see only minor changes to our forecast.\n",
    "\n",
    "For simplicity, in this exercise we will not be re-estimating the model parameters with the updated data, although that is certainly possible.\n",
    "\n",
    "There are a variety of methods in the results object that make it easy to extend the model with new data or even apply a given model to an entirely different dataset. They are `append`, `extend`, and `apply`. For more details about these methods, see [this example notebook](https://www.statsmodels.org/devel/examples/notebooks/generated/statespace_forecasting.html#Cross-validation).\n",
    "\n",
    "Here, we will use the `apply` method, which applies the model and estimated parameters to a new dataset.\n",
    "\n",
    "**Notes**:\n",
    "\n",
    "1. If `standardize=True` was used in model creation (which is the default), then the `apply` method will use the same standardization on the new dataset as on the original dataset by default. This is important when exploring the impacts of the \"news\", as we will be doing, and it is another reason that it is usually easiest to leave standardization to the model (if you prefer to re-standardize the new dataset, you can use the `retain_standardization=False` argument to the `apply` method).\n",
    "2. Because of the fact that some data in the FRED-MD dataset is revised after its initial publication, we are not just adding new observations but are potentially changing the values of previously observed entries (this is why we need to use the `apply` method rather than the `append` method)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Since we will be collecting results for a number of vintages,\n",
    "# construct a dictionary to hold them, and include the baseline\n",
    "# results from February 2020\n",
    "vintage_results = {'2020-02': results}\n",
    "\n",
    "# Get the updated monthly and quarterly datasets\n",
    "start = '2000'\n",
    "updated_endog_m = dta['2020-03'].dta_m.loc[start:, :]\n",
    "gdp_description = defn_q.loc['GDPC1', 'description']\n",
    "updated_endog_q = dta['2020-03'].dta_q.loc[start:, [gdp_description]]\n",
    "\n",
    "# Get the results for March 2020 using `apply`\n",
    "vintage_results['2020-03'] = results.apply(\n",
    "    updated_endog_m, endog_quarterly=updated_endog_q)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This new results object has all of the same attributes and methods as the baseline results object. For example, we can compute the updated forecast for real GDP growth in 2020Q2."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "March 2020 forecast for real GDP growth in 2020Q2: 2.52%\n"
     ]
    }
   ],
   "source": [
    "# Print the updated forecast for real GDP growth in 2020Q2\n",
    "updated_forecasts_q = (\n",
    "    vintage_results['2020-03'].forecast('June 2020')[gdp_description]\n",
    "                              .resample('Q').last())\n",
    "\n",
    "print('March 2020 forecast for real GDP growth in 2020Q2:'\n",
    "      f' {updated_forecasts_q[\"2020Q2\"]:.2f}%')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As expected, the forecast from March 2020 is only a little changed from our baseline (February 2020) forecast.\n",
    "\n",
    "We can continue this process, however, for the April, May, and June vintages and see how the incoming economic data changes our forecast for real GDP."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Apply our results to the remaining vintages\n",
    "for vintage in ['2020-04', '2020-05', '2020-06']:\n",
    "    # Get updated data for the vintage\n",
    "    updated_endog_m = dta[vintage].dta_m.loc[start:, :]\n",
    "    updated_endog_q = dta[vintage].dta_q.loc[start:, [gdp_description]]\n",
    "\n",
    "    # Get updated results for for the vintage\n",
    "    vintage_results[vintage] = results.apply(\n",
    "        updated_endog_m, endog_quarterly=updated_endog_q)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2020-02 forecast for real GDP growth in 2020Q2: 2.68%\n",
      "2020-03 forecast for real GDP growth in 2020Q2: 2.52%\n",
      "2020-04 forecast for real GDP growth in 2020Q2: -8.23%\n",
      "2020-05 forecast for real GDP growth in 2020Q2: -37.17%\n",
      "2020-06 forecast for real GDP growth in 2020Q2: -22.33%\n"
     ]
    }
   ],
   "source": [
    "# Compute forecasts for each vintage\n",
    "forecasts = {vintage: res.forecast('June 2020')[gdp_description]\n",
    "                         .resample('Q').last().loc['June 2020']\n",
    "             for vintage, res in vintage_results.items()}\n",
    "# Convert to a Pandas series with a date index\n",
    "forecasts = pd.Series(list(forecasts.values()),\n",
    "                      index=pd.PeriodIndex(forecasts.keys(), freq='M'))\n",
    "    \n",
    "# Print our forecast for 2020Q2 real GDP growth across all vintages\n",
    "for vintage, value in forecasts.items():\n",
    "    print(f'{vintage} forecast for real GDP growth in 2020Q2:'\n",
    "          f' {value:.2f}%')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Although there was not much of a change in the forecast between the February and March vintages, the forecasts from the April and May vintages each saw substantial declines.\n",
    "\n",
    "To dig into why the forecast changed so much, we can compute the impacts on the forecast of each piece of new information in each of the data updates. This computation of the \"news\" and its impact on forecasts follows Bańbura and Modugno (2014).\n",
    "\n",
    "Computation of the \"news\" and the associated impact is straightforward using the `news` method of the results object associated with one of the vintages. The basic syntax is:\n",
    "\n",
    "```python\n",
    "results.news(previous_vintage_results, impact_date='2020-06',\n",
    "             impacted_variable=gdp_description,\n",
    "             comparison_type='previous')\n",
    "```\n",
    "\n",
    "The \"news\" is then unexpected component of the updated datapoints in `results` that were not present in the `previous_vintage_results`, and the impacts will be computed for forecasts related to June 2020 (recall that for the mixed frequency setup here, the quarterly values are identified with the last monthly value in each quarter). This method returns a new results object, with a number of tables that decompose the impacts in variety of ways. For additional details about the computation of the \"news\" and the associated impacts, see [this example notebook](https://www.statsmodels.org/devel/examples/notebooks/generated/statespace_news.html).\n",
    "\n",
    "As an example, we will examine the news and impacts associated with the April 2020 vintage, compared to the March 2020 vintage."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "# Compute the news and impacts on the real GDP growth forecast\n",
    "# for 2020Q2, between the April and March vintages\n",
    "news = vintage_results['2020-04'].news(\n",
    "    vintage_results['2020-03'], impact_date='2020-06',\n",
    "    impacted_variable=gdp_description,\n",
    "    comparison_type='previous')\n",
    "\n",
    "# The `summary` method summarizes all updates. Here we aren't\n",
    "# showing it, to save space.\n",
    "# news.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Details of impacts: April vintage compared to March vintage**\n",
    "\n",
    "Now, we will show which ten new observations had the largest impact (in absolute value) on the forecast of real GDP growth in 2020Q2. This is shown in the table below, which has seven columns:\n",
    "\n",
    "- The first column, \"update date\", is the date of the new observation.\n",
    "- The second column, \"updated variable\", is the variable updated\n",
    "- The third column, \"observed\", shows the actual recorded value\n",
    "- The fourth column, \"forecast (prev)\", shows what value had been expected in the previous vintage (here, in the March 2020 vintage).\n",
    "- The fifth column, \"news\", shows the unexpected component of the update (it is equal to observed - forecast (prev))\n",
    "- The sixth column, \"weight\", shows how much weight this date / variable combination has on the forecast of interest\n",
    "- The final column, \"impact\", shows how much the forecast of real GDP growth in 2020Q2 changed based only on the single new observation captured by each given row\n",
    "\n",
    "From this table, we can see that in the April vintage, the largest impacts on the real GDP forecast for 2020Q2 came from:\n",
    "\n",
    "- Initial unemployment claims and the CBOE S&P 100 Volatility Index (VXO) each came in much higher than expected\n",
    "- Corporate bond spreads (AAA and BAA) came in higher than expected\n",
    "- Industrial production (including final products, manufacturing, durable materials, and the overall index) and capcity utilization came in much lower than expected"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>observed</th>\n",
       "      <th>forecast (prev)</th>\n",
       "      <th>news</th>\n",
       "      <th>weight</th>\n",
       "      <th>impact</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>update date</th>\n",
       "      <th>updated variable</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"10\" valign=\"top\">2020-03</th>\n",
       "      <th>IP: Final Products and Nonindustrial Supplies</th>\n",
       "      <td>-6.53</td>\n",
       "      <td>-0.11</td>\n",
       "      <td>-6.42</td>\n",
       "      <td>0.61</td>\n",
       "      <td>-3.92</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Initial Claims</th>\n",
       "      <td>253.49</td>\n",
       "      <td>0.94</td>\n",
       "      <td>252.54</td>\n",
       "      <td>-0.01</td>\n",
       "      <td>-2.06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Moody’s Baa Corporate Bond Minus FEDFUNDS</th>\n",
       "      <td>3.66</td>\n",
       "      <td>2.10</td>\n",
       "      <td>1.56</td>\n",
       "      <td>-0.63</td>\n",
       "      <td>-0.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>IP: Final Products (Market Group)</th>\n",
       "      <td>-6.52</td>\n",
       "      <td>-0.12</td>\n",
       "      <td>-6.39</td>\n",
       "      <td>-0.13</td>\n",
       "      <td>0.84</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>VXO</th>\n",
       "      <td>63.88</td>\n",
       "      <td>19.53</td>\n",
       "      <td>44.35</td>\n",
       "      <td>-0.01</td>\n",
       "      <td>-0.60</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>IP: Manufacturing (SIC)</th>\n",
       "      <td>-6.46</td>\n",
       "      <td>0.05</td>\n",
       "      <td>-6.50</td>\n",
       "      <td>0.06</td>\n",
       "      <td>-0.40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Moody’s Aaa Corporate Bond Minus FEDFUNDS</th>\n",
       "      <td>2.39</td>\n",
       "      <td>1.25</td>\n",
       "      <td>1.14</td>\n",
       "      <td>-0.35</td>\n",
       "      <td>-0.40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>IP: Consumer Goods</th>\n",
       "      <td>-6.06</td>\n",
       "      <td>-0.21</td>\n",
       "      <td>-5.84</td>\n",
       "      <td>-0.07</td>\n",
       "      <td>0.39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All Employees: Total nonfarm</th>\n",
       "      <td>-0.46</td>\n",
       "      <td>0.15</td>\n",
       "      <td>-0.61</td>\n",
       "      <td>0.53</td>\n",
       "      <td>-0.32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6-Month Treasury Bill:</th>\n",
       "      <td>-1.18</td>\n",
       "      <td>-0.06</td>\n",
       "      <td>-1.12</td>\n",
       "      <td>0.27</td>\n",
       "      <td>-0.30</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                           observed  \\\n",
       "update date updated variable                                          \n",
       "2020-03     IP: Final Products and Nonindustrial Supplies     -6.53   \n",
       "            Initial Claims                                   253.49   \n",
       "            Moody’s Baa Corporate Bond Minus FEDFUNDS          3.66   \n",
       "            IP: Final Products (Market Group)                 -6.52   \n",
       "            VXO                                               63.88   \n",
       "            IP: Manufacturing (SIC)                           -6.46   \n",
       "            Moody’s Aaa Corporate Bond Minus FEDFUNDS          2.39   \n",
       "            IP: Consumer Goods                                -6.06   \n",
       "            All Employees: Total nonfarm                      -0.46   \n",
       "            6-Month Treasury Bill:                            -1.18   \n",
       "\n",
       "                                                           forecast (prev)  \\\n",
       "update date updated variable                                                 \n",
       "2020-03     IP: Final Products and Nonindustrial Supplies            -0.11   \n",
       "            Initial Claims                                            0.94   \n",
       "            Moody’s Baa Corporate Bond Minus FEDFUNDS                 2.10   \n",
       "            IP: Final Products (Market Group)                        -0.12   \n",
       "            VXO                                                      19.53   \n",
       "            IP: Manufacturing (SIC)                                   0.05   \n",
       "            Moody’s Aaa Corporate Bond Minus FEDFUNDS                 1.25   \n",
       "            IP: Consumer Goods                                       -0.21   \n",
       "            All Employees: Total nonfarm                              0.15   \n",
       "            6-Month Treasury Bill:                                   -0.06   \n",
       "\n",
       "                                                             news  weight  \\\n",
       "update date updated variable                                                \n",
       "2020-03     IP: Final Products and Nonindustrial Supplies   -6.42    0.61   \n",
       "            Initial Claims                                 252.54   -0.01   \n",
       "            Moody’s Baa Corporate Bond Minus FEDFUNDS        1.56   -0.63   \n",
       "            IP: Final Products (Market Group)               -6.39   -0.13   \n",
       "            VXO                                             44.35   -0.01   \n",
       "            IP: Manufacturing (SIC)                         -6.50    0.06   \n",
       "            Moody’s Aaa Corporate Bond Minus FEDFUNDS        1.14   -0.35   \n",
       "            IP: Consumer Goods                              -5.84   -0.07   \n",
       "            All Employees: Total nonfarm                    -0.61    0.53   \n",
       "            6-Month Treasury Bill:                          -1.12    0.27   \n",
       "\n",
       "                                                           impact  \n",
       "update date updated variable                                       \n",
       "2020-03     IP: Final Products and Nonindustrial Supplies   -3.92  \n",
       "            Initial Claims                                  -2.06  \n",
       "            Moody’s Baa Corporate Bond Minus FEDFUNDS       -0.99  \n",
       "            IP: Final Products (Market Group)                0.84  \n",
       "            VXO                                             -0.60  \n",
       "            IP: Manufacturing (SIC)                         -0.40  \n",
       "            Moody’s Aaa Corporate Bond Minus FEDFUNDS       -0.40  \n",
       "            IP: Consumer Goods                               0.39  \n",
       "            All Employees: Total nonfarm                    -0.32  \n",
       "            6-Month Treasury Bill:                          -0.30  "
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# We can re-arrange the `details_by_impact` table to show the new\n",
    "# observations with the top ten impacts (in absolute value)\n",
    "details = news.details_by_impact\n",
    "details.index = details.index.droplevel(['impact date', 'impacted variable'])\n",
    "details['absolute impact'] = np.abs(details['impact'])\n",
    "details = (details.sort_values('absolute impact', ascending=False)\n",
    "                  .drop('absolute impact', axis=1))\n",
    "details.iloc[:10].round(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For each updated vintage of data, we can compute the news in the same way. Below, we compute all news vintages."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "news_results = {}\n",
    "vintages = ['2020-02', '2020-03', '2020-04', '2020-05', '2020-06']\n",
    "impact_date = '2020-06'\n",
    "\n",
    "for i in range(1, len(vintages)):\n",
    "    vintage = vintages[i]\n",
    "    prev_vintage = vintages[i - 1]\n",
    "\n",
    "    # Notice that to get the \"incremental\" news, we are computing\n",
    "    # the news relative to the previous vintage and not to the baseline\n",
    "    # (February 2020) vintage\n",
    "    news_results[vintage] = vintage_results[vintage].news(\n",
    "        vintage_results[prev_vintage],\n",
    "        impact_date=impact_date,\n",
    "        impacted_variable=gdp_description,\n",
    "        comparison_type='previous')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Impacts by group: evolution across all vintages**\n",
    "\n",
    "To summarize the news, we will take an approach similar to that of the [New York Fed Staff Nowcast](https://www.newyorkfed.org/research/policy/nowcast.html), and combine impacts by the groups defined above (for example \"Output and Income\", etc.).\n",
    "\n",
    "**Note**: the [New York Fed Staff Nowcast](https://www.newyorkfed.org/research/policy/nowcast.html) uses precisely the same dynamic factor model and estimation rountine (EM algorithm) to compute their nowcast, although they use a different dataset and different factor specification. In addition, they update their dataset and forecast every week, while the FRED-MD dataset we're using here only updates every month."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>2020-03</th>\n",
       "      <th>2020-04</th>\n",
       "      <th>2020-05</th>\n",
       "      <th>2020-06</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>group</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Output and Income</th>\n",
       "      <td>-0.12</td>\n",
       "      <td>-3.85</td>\n",
       "      <td>-10.08</td>\n",
       "      <td>9.15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Labor Market</th>\n",
       "      <td>0.29</td>\n",
       "      <td>-3.38</td>\n",
       "      <td>-19.76</td>\n",
       "      <td>5.94</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Housing</th>\n",
       "      <td>-0.09</td>\n",
       "      <td>-0.05</td>\n",
       "      <td>-0.00</td>\n",
       "      <td>0.26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Consumption, orders, and inventories</th>\n",
       "      <td>-0.04</td>\n",
       "      <td>-0.39</td>\n",
       "      <td>-0.35</td>\n",
       "      <td>0.47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Money and credit</th>\n",
       "      <td>-0.02</td>\n",
       "      <td>-0.01</td>\n",
       "      <td>-0.08</td>\n",
       "      <td>0.05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Interest and exchange rates</th>\n",
       "      <td>-0.13</td>\n",
       "      <td>-2.67</td>\n",
       "      <td>0.94</td>\n",
       "      <td>0.97</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Prices</th>\n",
       "      <td>0.00</td>\n",
       "      <td>0.01</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Stock market</th>\n",
       "      <td>-0.05</td>\n",
       "      <td>-0.35</td>\n",
       "      <td>0.02</td>\n",
       "      <td>0.33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Total impact on 2020Q2 forecast</th>\n",
       "      <td>-0.16</td>\n",
       "      <td>-10.70</td>\n",
       "      <td>-29.30</td>\n",
       "      <td>17.18</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                      2020-03  2020-04  2020-05  2020-06\n",
       "group                                                                   \n",
       "Output and Income                       -0.12    -3.85   -10.08     9.15\n",
       "Labor Market                             0.29    -3.38   -19.76     5.94\n",
       "Housing                                 -0.09    -0.05    -0.00     0.26\n",
       "Consumption, orders, and inventories    -0.04    -0.39    -0.35     0.47\n",
       "Money and credit                        -0.02    -0.01    -0.08     0.05\n",
       "Interest and exchange rates             -0.13    -2.67     0.94     0.97\n",
       "Prices                                   0.00     0.01     0.00     0.00\n",
       "Stock market                            -0.05    -0.35     0.02     0.33\n",
       "Total impact on 2020Q2 forecast         -0.16   -10.70   -29.30    17.18"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "group_impacts = {'2020-02': None}\n",
    "\n",
    "for vintage, news in news_results.items():\n",
    "    # Start from the details by impact table\n",
    "    details_by_impact = (\n",
    "        news.details_by_impact.reset_index()\n",
    "            .drop(['impact date', 'impacted variable'], axis=1))\n",
    "    \n",
    "    # Merge with the groups dataset, so that we can identify\n",
    "    # which group each individual impact belongs to\n",
    "    impacts = (pd.merge(details_by_impact, groups, how='left',\n",
    "                        left_on='updated variable', right_on='description')\n",
    "                 .drop('description', axis=1)\n",
    "                 .set_index(['update date', 'updated variable']))\n",
    "\n",
    "    # Compute impacts by group, summing across the individual impacts\n",
    "    group_impacts[vintage] = impacts.groupby('group').sum()['impact']\n",
    "\n",
    "# Add in a row of zeros for the baseline forecast\n",
    "group_impacts['2020-02'] = group_impacts['2020-03'] * np.nan\n",
    "\n",
    "# Convert into a Pandas DataFrame, and fill in missing entries\n",
    "# with zeros (missing entries happen when there were no updates\n",
    "# for a given group in a given vintage)\n",
    "group_impacts = (\n",
    "    pd.concat(group_impacts, axis=1)\n",
    "      .fillna(0)\n",
    "      .reindex(group_counts.index).T)\n",
    "group_impacts.index = forecasts.index\n",
    "\n",
    "# Print the table of impacts from data in each group,\n",
    "# along with a row with the \"Total\" impact\n",
    "(group_impacts.T\n",
    "    .append(group_impacts.sum(axis=1).rename('Total impact on 2020Q2 forecast'))\n",
    "    .round(2).iloc[:, 1:])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Impacts by group: graphical illustration**\n",
    "\n",
    "While the table is informative, a graphical version can be even more helpful. Below, we show a figure of the type shown in Bańbura and Modugno (2014), but also used in, for example, the [New York Fed Staff Nowcast](https://www.newyorkfed.org/research/policy/nowcast.html)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "with sns.color_palette('deep'):\n",
    "    fig, ax = plt.subplots(figsize=(14, 6))\n",
    "\n",
    "    # Stacked bar plot showing the impacts by group\n",
    "    group_impacts.plot(kind='bar', stacked=True, width=0.3, zorder=2, ax=ax);\n",
    "\n",
    "    # Line plot showing the forecast for real GDP growth in 2020Q2 for each vintage\n",
    "    x = np.arange(len(forecasts))\n",
    "    ax.plot(x, forecasts, marker='o', color='k', markersize=7, linewidth=2)\n",
    "    ax.hlines(0, -1, len(group_impacts) + 1, linewidth=1)\n",
    "\n",
    "    # x-ticks\n",
    "    labels = group_impacts.index.strftime('%b')\n",
    "    ax.xaxis.set_ticklabels(labels)\n",
    "    ax.xaxis.set_tick_params(size=0)\n",
    "    ax.xaxis.set_tick_params(labelrotation='auto', labelsize=13)\n",
    "\n",
    "    # y-ticks\n",
    "    ax.yaxis.set_tick_params(direction='in', size=0, labelsize=13)\n",
    "    ax.yaxis.grid(zorder=0)\n",
    "    \n",
    "    # title, remove spines\n",
    "    ax.set_title('Evolution of real GDP growth nowcast: 2020Q2', fontsize=16, fontweight=600, loc='left')\n",
    "    [ax.spines[spine].set_visible(False)\n",
    "     for spine in ['top', 'left', 'bottom', 'right']]\n",
    "    \n",
    "    # base forecast vs updates\n",
    "    ylim = ax.get_ylim()\n",
    "    ax.vlines(0.5, ylim[0], ylim[1] + 5, linestyles='--')\n",
    "    ax.annotate('Base forecast', (-0.2, 22), fontsize=14)\n",
    "    ax.annotate(r'Updated forecasts and impacts from the \"news\" $\\rightarrow$', (0.65, 22), fontsize=14)\n",
    "\n",
    "    # legend\n",
    "    ax.legend(loc='upper center', ncol=4, fontsize=13, bbox_to_anchor=(0.5, -0.1), frameon=False)\n",
    "\n",
    "    fig.tight_layout();"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### References\n",
    "\n",
    "Bańbura, Marta, and Michele Modugno. \"Maximum likelihood estimation of factor models on datasets with arbitrary pattern of missing data.\" Journal of Applied Econometrics 29, no. 1 (2014): 133-160.\n",
    "\n",
    "Bańbura, Marta, Domenico Giannone, and Lucrezia Reichlin. \"Nowcasting.\" The Oxford Handbook of Economic Forecasting. July 8, 2011.\n",
    "\n",
    "Bok, Brandyn, Daniele Caratelli, Domenico Giannone, Argia M. Sbordone, and Andrea Tambalotti. 2018. \"Macroeconomic Nowcasting and Forecasting with Big Data.\" Annual Review of Economics 10 (1): 615-43.\n",
    "\n",
    "Mariano, Roberto S., and Yasutomo Murasawa. \"A coincident index, common factors, and monthly real GDP.\" Oxford Bulletin of Economics and Statistics 72, no. 1 (2010): 27-46.\n",
    "\n",
    "McCracken, Michael W., and Serena Ng. \"FRED-MD: A monthly database for macroeconomic research.\" Journal of Business & Economic Statistics 34, no. 4 (2016): 574-589."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}

Great application and very useful notebook. In the 0.12 version of statsmodels will it also be possible to work with other frequencies (weekly/monthly: DynamicFactorWM? ) in the same way as shown for monthly/quarterly?

Thanks! For v0.12, it will only be monthly/quarterly. In the future, we will hopefully add support for additional mixed frequency combinations.

Very useful notebook. Weekly/Monthly frequencies will be super helpful. Thanks!

Thanks, I'm glad it's helpful. Yes, I hope that we can add additional frequencies soon.

Hi, I can not seem to open the url https://s3.amazonaws.com/files.fred.stlouisfed.org/fred-md. When running this notebook, I get: HTTP Error 403: Forbidden. Am I doing something wrong? Thanks in advance!

Change base_url = 'https://files.stlouisfed.org/files/htdocs/fred-md'

Hello, I'm trying to run the notebook, but I ran into an issue with the 'fredmd_definitions.csv' and 'fredqd_definitions.csv' files. I tried to access the URL in the notebook, but I was met with an error message. I then went to FRED website and downloaded the csv files, but they seem to have different column names, which is creating issues with the grouping.

Would you be able to share the csv appendix files used for the notebook.

Thank You!

hi, does NowCasting model predict a quarter-on-quarter growth rate of GDP? If it is a quarter-on-quarter growth rate, can it be predicted, or can it be calculated?

Hi,
Thank you very much for sharing your work. I wonder if the DynamicFactorMQ can also work with the mixed frequency data between yearly and quarterly? And more importantly, if this package can also work with panel data (i.e. multiple countries with data observed across years). Many thanks again!

Hi,
Thank you very much for your work. I wonder how DynamicFactorMQ deals with missing data when you have non-synchronized release indexes. I couldn't find documentation about this issue. Could you please tell me a little more about this?

Hi @jeaniek, yes, you could use DynamicFactorMQ with quarterly / annual data, and with multiple countries and multiple years. NOte that the package does not do anything special in those cases; it would just be a typical dynamic factor model.

Hi @lladamartin, I'm not sure I understand exactly what you're asking. Generally, the model allows for arbitrary patterns of missing observations. For example, if you data through 2024Q1 for one dataset, while another dataset has only been released through 2023Q4, then you would have a NaN value for the second dataset for 2024Q1. The model would run just fine in this case.

Thank you for your response. I apologize for the lack of clarity in my question. I have monthly information on a set of indicators. These indicators have 1, 2, and up to 3 months of lag. I want to obtain the latent factor of this information set. How does the model estimate the value of the factor in April if many indicators have missing data for that month and previous months? How does the model handle those missing values?

Because it is a state space model, where the unobserved state has a defined transition equation, it can produce an estimate for the factor in April even if you had no data for the month (i.e. it just estimates April using its estimate for March combined with the definition of how the state transitions between periods). As you start to observe parts of the data for April, it updates its estimate using whatever data is available. A more detailed description of how this works can be found in, e.g., Maximum likelihood estimation of factor models on datasets with arbitrary pattern of missing data

I'll see the paper! Thanks!