TSLS.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Generate some data using the left graph."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "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>Z</th>\n",
       "      <th>X</th>\n",
       "      <th>Y</th>\n",
       "      <th>W</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-1.088058</td>\n",
       "      <td>-10.024597</td>\n",
       "      <td>-24.805915</td>\n",
       "      <td>-2.081504</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-0.254004</td>\n",
       "      <td>-6.036450</td>\n",
       "      <td>-18.310224</td>\n",
       "      <td>-1.301719</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.238271</td>\n",
       "      <td>0.873657</td>\n",
       "      <td>2.892484</td>\n",
       "      <td>-0.315459</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.566307</td>\n",
       "      <td>7.403453</td>\n",
       "      <td>19.262975</td>\n",
       "      <td>0.820820</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>-1.814836</td>\n",
       "      <td>-12.857574</td>\n",
       "      <td>-34.418315</td>\n",
       "      <td>-3.197533</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Z          X          Y         W\n",
       "0 -1.088058 -10.024597 -24.805915 -2.081504\n",
       "1 -0.254004  -6.036450 -18.310224 -1.301719\n",
       "2  0.238271   0.873657   2.892484 -0.315459\n",
       "3  0.566307   7.403453  19.262975  0.820820\n",
       "4 -1.814836 -12.857574 -34.418315 -3.197533"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "N = 1000 # sample size\n",
    "\n",
    "B = 2. # \"causal effect\" of X on Y\n",
    "\n",
    "z = np.random.normal(size=N)\n",
    "u = np.random.normal(size=N)\n",
    "w = np.random.normal(z)\n",
    "x = np.random.normal(3.*z + 2.*u + 3.*w)\n",
    "y = np.random.normal(B*x + 2.*u + 3.*w)\n",
    "\n",
    "df = pd.DataFrame({'Z': z, 'X': x, 'Y': y, 'W': w})\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### What does the data look like?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 16 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "pd.plotting.scatter_matrix(df, figsize=(8,8));"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### What is the (incorrect) naive effect?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>            <td>Y</td>        <th>  R-squared:         </th> <td>   0.989</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.989</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>9.133e+04</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Sat, 07 Dec 2019</td> <th>  Prob (F-statistic):</th>  <td>  0.00</td>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>15:39:43</td>     <th>  Log-Likelihood:    </th> <td> -2061.9</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>  1000</td>      <th>  AIC:               </th> <td>   4126.</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>   999</td>      <th>  BIC:               </th> <td>   4131.</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     1</td>      <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>    \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <td></td>     <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>X</th> <td>    2.6245</td> <td>    0.009</td> <td>  302.202</td> <td> 0.000</td> <td>    2.607</td> <td>    2.642</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 5.384</td> <th>  Durbin-Watson:     </th> <td>   1.948</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.068</td> <th>  Jarque-Bera (JB):  </th> <td>   5.363</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td>-0.179</td> <th>  Prob(JB):          </th> <td>  0.0685</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 3.008</td> <th>  Cond. No.          </th> <td>    1.00</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:                      Y   R-squared:                       0.989\n",
       "Model:                            OLS   Adj. R-squared:                  0.989\n",
       "Method:                 Least Squares   F-statistic:                 9.133e+04\n",
       "Date:                Sat, 07 Dec 2019   Prob (F-statistic):               0.00\n",
       "Time:                        15:39:43   Log-Likelihood:                -2061.9\n",
       "No. Observations:                1000   AIC:                             4126.\n",
       "Df Residuals:                     999   BIC:                             4131.\n",
       "Df Model:                           1                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "==============================================================================\n",
       "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "X              2.6245      0.009    302.202      0.000       2.607       2.642\n",
       "==============================================================================\n",
       "Omnibus:                        5.384   Durbin-Watson:                   1.948\n",
       "Prob(Omnibus):                  0.068   Jarque-Bera (JB):                5.363\n",
       "Skew:                          -0.179   Prob(JB):                       0.0685\n",
       "Kurtosis:                       3.008   Cond. No.                         1.00\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from statsmodels.api import OLS\n",
    "\n",
    "model = OLS(exog=df[['X']], endog=df['Y'])\n",
    "result = model.fit()\n",
    "result.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### We measure 2.62, which is badly biased for B=2.0"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### What do we get with a two-stage estimator? Let's do it manually, just for fun.\n",
    "\n",
    "First stage: Z -> X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>            <td>X</td>        <th>  R-squared:         </th> <td>   0.899</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.899</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   4429.</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Sat, 07 Dec 2019</td> <th>  Prob (F-statistic):</th>  <td>  0.00</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>15:39:44</td>     <th>  Log-Likelihood:    </th> <td> -2209.7</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>  1000</td>      <th>  AIC:               </th> <td>   4423.</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>   998</td>      <th>  BIC:               </th> <td>   4433.</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     2</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <td></td>     <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Z</th> <td>    2.9428</td> <td>    0.101</td> <td>   29.152</td> <td> 0.000</td> <td>    2.745</td> <td>    3.141</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>W</th> <td>    3.0187</td> <td>    0.071</td> <td>   42.757</td> <td> 0.000</td> <td>    2.880</td> <td>    3.157</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 0.577</td> <th>  Durbin-Watson:     </th> <td>   2.039</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.749</td> <th>  Jarque-Bera (JB):  </th> <td>   0.463</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.036</td> <th>  Prob(JB):          </th> <td>   0.793</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 3.078</td> <th>  Cond. No.          </th> <td>    2.63</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:                      X   R-squared:                       0.899\n",
       "Model:                            OLS   Adj. R-squared:                  0.899\n",
       "Method:                 Least Squares   F-statistic:                     4429.\n",
       "Date:                Sat, 07 Dec 2019   Prob (F-statistic):               0.00\n",
       "Time:                        15:39:44   Log-Likelihood:                -2209.7\n",
       "No. Observations:                1000   AIC:                             4423.\n",
       "Df Residuals:                     998   BIC:                             4433.\n",
       "Df Model:                           2                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "==============================================================================\n",
       "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "Z              2.9428      0.101     29.152      0.000       2.745       3.141\n",
       "W              3.0187      0.071     42.757      0.000       2.880       3.157\n",
       "==============================================================================\n",
       "Omnibus:                        0.577   Durbin-Watson:                   2.039\n",
       "Prob(Omnibus):                  0.749   Jarque-Bera (JB):                0.463\n",
       "Skew:                           0.036   Prob(JB):                        0.793\n",
       "Kurtosis:                       3.078   Cond. No.                         2.63\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_stage_1 = OLS(exog=df[['Z', 'W']], endog=df['X'])\n",
    "result = model_stage_1.fit()\n",
    "result.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['E[X|Z, W]'] = result.predict()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Second stage:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>            <td>Y</td>        <th>  R-squared:         </th> <td>   0.880</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.879</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   3644.</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Sat, 07 Dec 2019</td> <th>  Prob (F-statistic):</th>  <td>  0.00</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>15:39:46</td>     <th>  Log-Likelihood:    </th> <td> -3266.8</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>  1000</td>      <th>  AIC:               </th> <td>   6538.</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>   998</td>      <th>  BIC:               </th> <td>   6547.</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     2</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>      <td>nonrobust</td>    <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "      <td></td>         <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>E[X|Z, W]</th> <td>    1.9951</td> <td>    0.099</td> <td>   20.210</td> <td> 0.000</td> <td>    1.801</td> <td>    2.189</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>W</th>         <td>    3.0074</td> <td>    0.464</td> <td>    6.477</td> <td> 0.000</td> <td>    2.096</td> <td>    3.919</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 1.062</td> <th>  Durbin-Watson:     </th> <td>   2.012</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.588</td> <th>  Jarque-Bera (JB):  </th> <td>   0.929</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.051</td> <th>  Prob(JB):          </th> <td>   0.629</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 3.109</td> <th>  Cond. No.          </th> <td>    15.8</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:                      Y   R-squared:                       0.880\n",
       "Model:                            OLS   Adj. R-squared:                  0.879\n",
       "Method:                 Least Squares   F-statistic:                     3644.\n",
       "Date:                Sat, 07 Dec 2019   Prob (F-statistic):               0.00\n",
       "Time:                        15:39:46   Log-Likelihood:                -3266.8\n",
       "No. Observations:                1000   AIC:                             6538.\n",
       "Df Residuals:                     998   BIC:                             6547.\n",
       "Df Model:                           2                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "==============================================================================\n",
       "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "E[X|Z, W]      1.9951      0.099     20.210      0.000       1.801       2.189\n",
       "W              3.0074      0.464      6.477      0.000       2.096       3.919\n",
       "==============================================================================\n",
       "Omnibus:                        1.062   Durbin-Watson:                   2.012\n",
       "Prob(Omnibus):                  0.588   Jarque-Bera (JB):                0.929\n",
       "Skew:                           0.051   Prob(JB):                        0.629\n",
       "Kurtosis:                       3.109   Cond. No.                         15.8\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model_stage_2 = OLS(exog=df[['E[X|Z, W]', 'W']], endog=df['Y'])\n",
    "result = model_stage_2.fit()\n",
    "result.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## So the coefficient in the second stage for the predictions on the first stage is about 2.0 (1.80 to 2.18 at the 95% CL), which is consistent with the true value of 2.0."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}