jarutis/Ekonometrija.ipynb

## Ekonometrija.ipynb
{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 5.8"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Calls per hour to a telephone exchange have a Poisson distribution\n",
    "$$p(X)=\\frac{\\lambda^{X}e^{-\\lambda}}{X!}\\textrm{,}\\qquad\\textrm{ with }E(x)=\\lambda$$\n",
    "During n randomly selected hours, the numbers of calls received by the exchanges are found to be $(X_1, X_2, X_3, ..., X_n)$. Find MLE of $\\lambda$."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Solution:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$$L(\\lambda)=\\prod_{i=1}^n\\frac{\\lambda^{X_i}e^{-\\lambda}}{X_i!}=e^{-n\\lambda}\\lambda^{\\sum_{i=1}^nX_i}\\prod_{i=1}^n\\frac{1}{X_i}$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$$l(\\lambda)=\\ln(L(\\lambda))=-n\\lambda + \\sum_{i=1}^nX_i \\ln(\\lambda) + \\sum_{i=1}^n(\\ln(1) - \\ln(X_i!))$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$$l'(\\lambda)=0 \\implies -n +\\sum_{i-1}^n X_i \\lambda^{-1} = 0 \\implies \\lambda = \\frac{\\sum_{i-1}^n X_i}{n}=\\bar{X}$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 9.11"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Consider the estimated consumption function\n",
    "\n",
    "$$\\hat{Y}=15 + 0.8 X_2 + 0.5 X_3$$\n",
    "\n",
    "where $Y = \\textrm{consumption}$, $X_2 = \\textrm{income}$ and $X_3 = \\textrm{wealth}$. If $n=20$, $\\sum e^2=34$ and, when variables are measured in deviation form,\n",
    "\n",
    "$$(x'x)^{-1}= \n",
    " \\begin{pmatrix}\n",
    "  0.02 & 0.005\\\\\n",
    "  0.005 & 0.03\n",
    " \\end{pmatrix}\n",
    ",$$\n",
    "\n",
    "test whether MPC income = MPC wealth."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Solution"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We need to test hypothesis $H_0 : \\beta_2=\\beta_3$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Test statistic $TS=\\frac{\\hat{\\beta_2}-\\hat{\\beta_3}}{u}$ has a Student's t distribution with $20-3=17$ degrees of freedom, where $u^2=s^2(x^{22}+x^{33}-2x^{23})$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$$s^2=\\frac{\\sum e^2}{n-3}=\\frac{34}{17}=2$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$$u^2=s^2(x^{22}+x^{33}-2x^{23})=2(0.02+0.03 -2\\cdot0.005)=0.08$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$$TS=\\frac{\\hat{\\beta_2}-\\hat{\\beta_3}}{u}\\approx 1.06$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Critical value:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "2.10981557783332"
      ],
      "text/latex": [
       "2.10981557783332"
      ],
      "text/markdown": [
       "2.10981557783332"
      ],
      "text/plain": [
       "[1] 2.109816"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "qt(1-0.025, 17)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Conclusion:** $H_0$ can not be rejected."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Computer exercise II (368p.)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Data set 4 from the book contains quarterly data on UK consumption and disposable income for the period 1974Q1 - 1984Q4."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "library(ggplot2)\n",
    "library(dplyr)\n",
    "\n",
    "\n",
    "library(repr)\n",
    "options(repr.plot.width=6, repr.plot.height=4)\n",
    "\n",
    "plot_theme <- theme_minimal() +\n",
    "  theme(\n",
    "    panel.border = element_rect(colour = \"black\", fill=NA, size=1),\n",
    "    legend.background = element_rect(fill=\"white\", linetype=\"solid\"),\n",
    "    legend.position = c(0.16, 0.85)\n",
    "  )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "Y <- c(48856, 48921, 50727, 50929, 51138, 49695, 50061, 49525, 49371, 49445, 51344, 49539, 47278, 47895, 48996, 51164, 50232, 52379, 53388, 54314, 53412, 55033, 55014, 58841, 55814, 56151, 56968, 56949, 56763, 55902, 55520, 55962, 55325, 55568, 55919, 56232, 55950, 57203, 57563, 58234, 57961, 58954, 58975, 61041)\n",
    "C <- c(42065, 43636, 44994, 47521, 42667, 43979, 44492, 46362, 42364, 43657, 44940, 47318, 42264, 42981, 44484, 47754, 44845, 45560, 47574, 49531, 46356, 49318, 48664, 51326, 48272, 47527, 49480, 50546, 47483, 47892, 49457, 51179, 47360, 47842, 50190, 52588, 49438, 50081, 52825, 54561, 50653, 51526, 52830, 55950)\n",
    "\n",
    "data <- data.frame(\n",
    "  Period = zoo::as.yearqtr(seq(as.Date(\"1974-01-01\"), as.Date(\"1984-12-31\"), by = \"quarter\")),\n",
    "  Y = Y,\n",
    "  C = C,\n",
    "  y = log(Y),\n",
    "  c = log(C),\n",
    "  D2 = rep(c(0, 1, 0, 0), 11),\n",
    "  D3 = rep(c(0, 0, 1, 0), 11),\n",
    "  D4 = rep(c(0, 0, 0, 1), 11)) %>%\n",
    "  mutate(c_l1 = lag(c, 1)) %>%\n",
    "  mutate(c_l2 = lag(c, 2)) %>%\n",
    "  mutate(c_l3 = lag(c, 3)) %>%\n",
    "  mutate(c_l4 = lag(c, 4)) %>%\n",
    "  mutate(y_l1 = lag(y, 1)) %>%\n",
    "  mutate(y_l2 = lag(y, 2)) %>%\n",
    "  mutate(y_l3 = lag(y, 3)) %>%\n",
    "  mutate(y_l4 = lag(y, 4))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {},
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtAAAAHgCAYAAACMxVqsAAAEGWlDQ1BrQ0dDb2xvclNwYWNl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\nYII=",
      "text/plain": [
       "plot without title"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ggplot(tidyr::gather(data, Series, Value, C, Y), aes(Period, Value, group=Series, linetype=Series)) + \n",
    "  geom_line() + \n",
    "  ylim(40000, 62500) +\n",
    "  ggtitle(\"UK consumption and disposable income\") +\n",
    "  labs(y = \"Value GBP billion\") +\n",
    "  zoo::scale_x_yearqtr(format=\"%YQ%q\", n=5) + \n",
    "  scale_linetype_discrete(breaks=c(\"Y\", \"C\"), labels=c(\"Disposable Income\", \"Consumption\")) +\n",
    "  plot_theme"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ADL model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Helper function for statistics used in the book\n",
    "statistics <- function(fit) {\n",
    "  print(summary(fit))\n",
    "  print(paste(\"Sum of squared residuals:\", sum(resid(fit)^2)))\n",
    "  print(lmtest::dwtest(fit))\n",
    "  print(lmtest::resettest(fit, power = c(2)))\n",
    "  print(tseries::jarque.bera.test(resid(fit)))\n",
    "  print(lmtest::bgtest(fit, order=4))\n",
    "  print(lmtest::bptest(fit)) # didn't find LM statistic for heteroskedasticity as used in the book, used Breush and Pagan test instead\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Call:\n",
      "lm(formula = model, data = sample)\n",
      "\n",
      "Residuals:\n",
      "      Min        1Q    Median        3Q       Max \n",
      "-0.029710 -0.005864 -0.001334  0.005772  0.042917 \n",
      "\n",
      "Coefficients:\n",
      "             Estimate Std. Error t value Pr(>|t|)    \n",
      "(Intercept)  0.014305   0.448594   0.032  0.97477    \n",
      "c_l1         0.498618   0.153227   3.254  0.00275 ** \n",
      "c_l2         0.416320   0.165454   2.516  0.01725 *  \n",
      "y            0.272000   0.134187   2.027  0.05133 .  \n",
      "y_l1         0.252028   0.152561   1.652  0.10863    \n",
      "y_l2        -0.446402   0.134436  -3.321  0.00231 ** \n",
      "D2           0.045384   0.017653   2.571  0.01516 *  \n",
      "D3           0.086441   0.011423   7.568 1.57e-08 ***\n",
      "D4           0.105978   0.009664  10.967 3.38e-12 ***\n",
      "---\n",
      "Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1\n",
      "\n",
      "Residual standard error: 0.01458 on 31 degrees of freedom\n",
      "Multiple R-squared:  0.9662,\tAdjusted R-squared:  0.9575 \n",
      "F-statistic: 110.9 on 8 and 31 DF,  p-value: < 2.2e-16\n",
      "\n",
      "[1] \"Sum of squared residuals: 0.00658799694266796\"\n",
      "\n",
      "\tDurbin-Watson test\n",
      "\n",
      "data:  fit\n",
      "DW = 1.8641, p-value = 0.3285\n",
      "alternative hypothesis: true autocorrelation is greater than 0\n",
      "\n",
      "\n",
      "\tRESET test\n",
      "\n",
      "data:  fit\n",
      "RESET = 0.31341, df1 = 1, df2 = 30, p-value = 0.5797\n",
      "\n",
      "\n",
      "\tJarque Bera Test\n",
      "\n",
      "data:  resid(fit)\n",
      "X-squared = 8.3655, df = 2, p-value = 0.01526\n",
      "\n",
      "\n",
      "\tBreusch-Godfrey test for serial correlation of order up to 4\n",
      "\n",
      "data:  fit\n",
      "LM test = 2.9251, df = 4, p-value = 0.5704\n",
      "\n",
      "\n",
      "\tstudentized Breusch-Pagan test\n",
      "\n",
      "data:  fit\n",
      "BP = 6.3187, df = 8, p-value = 0.6116\n",
      "\n"
     ]
    }
   ],
   "source": [
    "model <- c ~ c_l1 + c_l2 + y + y_l1 + y_l2 + D2 + D3 + D4\n",
    "sample <- data[as.character(data$Period) > \"1975\",]\n",
    "\n",
    "fit1247 <- lm(model, data = sample)\n",
    "statistics(fit1247)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Jarque Bera test indicates that the residuals might not follow a normal distribution."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Fourth-order lags"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As suggested we will fit a model with up to fourth-order lags:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_lag4 <- c ~ c_l1 + c_l2 + c_l3 + c_l4 + y + y_l1 + y_l2 + y_l3 + y_l4 + D2 + D3 + D4\n",
    "fit1247_lag4 <- lm(model_lag4, data = sample)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "0.00588903173591893"
      ],
      "text/latex": [
       "0.00588903173591893"
      ],
      "text/markdown": [
       "0.00588903173591893"
      ],
      "text/plain": [
       "[1] 0.005889032"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sum(resid(fit1247_lag4)^2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Adding higher order lags indeed decreases residuals slightly."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Call:\n",
      "lm(formula = model_lag4, data = sample)\n",
      "\n",
      "Residuals:\n",
      "      Min        1Q    Median        3Q       Max \n",
      "-0.027742 -0.005719 -0.001266  0.004391  0.043238 \n",
      "\n",
      "Coefficients:\n",
      "            Estimate Std. Error t value Pr(>|t|)    \n",
      "(Intercept) -0.14100    0.48159  -0.293 0.771928    \n",
      "c_l1         0.50940    0.18392   2.770 0.010025 *  \n",
      "c_l2         0.18521    0.23998   0.772 0.446940    \n",
      "c_l3         0.04458    0.21219   0.210 0.835161    \n",
      "c_l4         0.31957    0.18701   1.709 0.098954 .  \n",
      "y            0.34823    0.15255   2.283 0.030543 *  \n",
      "y_l1         0.15026    0.18456   0.814 0.422663    \n",
      "y_l2        -0.45244    0.17439  -2.594 0.015131 *  \n",
      "y_l3         0.13826    0.20433   0.677 0.504395    \n",
      "y_l4        -0.23327    0.17641  -1.322 0.197157    \n",
      "D2           0.04608    0.02173   2.121 0.043244 *  \n",
      "D3           0.05894    0.02628   2.242 0.033348 *  \n",
      "D4           0.07652    0.02038   3.754 0.000846 ***\n",
      "---\n",
      "Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1\n",
      "\n",
      "Residual standard error: 0.01477 on 27 degrees of freedom\n",
      "Multiple R-squared:  0.9698,\tAdjusted R-squared:  0.9564 \n",
      "F-statistic: 72.29 on 12 and 27 DF,  p-value: < 2.2e-16\n",
      "\n",
      "[1] \"Sum of squared residuals: 0.00588903173591893\"\n",
      "\n",
      "\tDurbin-Watson test\n",
      "\n",
      "data:  fit\n",
      "DW = 1.9511, p-value = 0.4136\n",
      "alternative hypothesis: true autocorrelation is greater than 0\n",
      "\n",
      "\n",
      "\tRESET test\n",
      "\n",
      "data:  fit\n",
      "RESET = 0.37595, df1 = 1, df2 = 26, p-value = 0.5451\n",
      "\n",
      "\n",
      "\tJarque Bera Test\n",
      "\n",
      "data:  resid(fit)\n",
      "X-squared = 17.742, df = 2, p-value = 0.0001404\n",
      "\n",
      "\n",
      "\tBreusch-Godfrey test for serial correlation of order up to 4\n",
      "\n",
      "data:  fit\n",
      "LM test = 3.6694, df = 4, p-value = 0.4526\n",
      "\n",
      "\n",
      "\tstudentized Breusch-Pagan test\n",
      "\n",
      "data:  fit\n",
      "BP = 8.3915, df = 12, p-value = 0.7538\n",
      "\n"
     ]
    }
   ],
   "source": [
    "statistics(fit1247_lag4)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Additional lagged variables do not seem to be significant and we still have normality issue indicated by Jarque Bera Test."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<thead><tr><th scope=col>Res.Df</th><th scope=col>RSS</th><th scope=col>Df</th><th scope=col>Sum of Sq</th><th scope=col>F</th><th scope=col>Pr(&gt;F)</th></tr></thead>\n",
       "<tbody>\n",
       "\t<tr><td>31          </td><td>0.006587997 </td><td>NA          </td><td>          NA</td><td>       NA   </td><td>       NA   </td></tr>\n",
       "\t<tr><td>27          </td><td>0.005889032 </td><td> 4          </td><td>0.0006989652</td><td>0.8011529   </td><td>0.5351064   </td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "\\begin{tabular}{r|llllll}\n",
       " Res.Df & RSS & Df & Sum of Sq & F & Pr(>F)\\\\\n",
       "\\hline\n",
       "\t 31           & 0.006587997  & NA           &           NA &        NA    &        NA   \\\\\n",
       "\t 27           & 0.005889032  &  4           & 0.0006989652 & 0.8011529    & 0.5351064   \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "Res.Df | RSS | Df | Sum of Sq | F | Pr(>F) | \n",
       "|---|---|\n",
       "| 31           | 0.006587997  | NA           |           NA |        NA    |        NA    | \n",
       "| 27           | 0.005889032  |  4           | 0.0006989652 | 0.8011529    | 0.5351064    | \n",
       "\n",
       "\n"
      ],
      "text/plain": [
       "  Res.Df RSS         Df Sum of Sq    F         Pr(>F)   \n",
       "1 31     0.006587997 NA           NA        NA        NA\n",
       "2 27     0.005889032  4 0.0006989652 0.8011529 0.5351064"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "car::linearHypothesis(fit1247_lag4,c(\"c_l3 = 0\", \"c_l4 = 0\",\"y_l3 = 0\", \"y_l4 = 0\"),test=\"F\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Residuals"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {},
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAtAAAAHgCAYAAACMxVqsAAAEGWlDQ1BrQ0dDb2xvclNwYWNl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\nCVCB9rV4WDkSIAESIAESIAESIAG/Efh/TZJjLG45zVkAAAAASUVORK5CYII=",
      "text/plain": [
       "plot without title"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "residuals <- resid(fit1247)\n",
    "residuals_df <- data.frame(residuals, Period = sample$Period)\n",
    "s <- sd(residuals)\n",
    "\n",
    "ggplot(residuals_df, aes(Period, residuals)) + \n",
    "  geom_line() +\n",
    "  labs(y = \"Residuals\", x = \"Period\") +\n",
    "  zoo::scale_x_yearqtr(format=\"%YQ%q\", n=5) +\n",
    "  geom_hline(yintercept=c(s, -s), linetype=\"dashed\") +\n",
    "  annotate(\"text\", x = 1975, y = s, label = \"standard deviation\", vjust = -1, hjust = 0.1, size = 3) + \n",
    "  annotate(\"text\", x = 1975, y = -s, label = \"standard deviation\", vjust = 2, hjust = 0.1, size = 3) + \n",
    "  plot_theme"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From the residuals plot two periods look like outliers, which might be the reason for failed normality test."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Log-linear version of the partial adjustment model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_pa <- c ~ c_l1 + y + D2 + D3 + D4\n",
    "fit1247_pa <- lm(model_pa, data = sample)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Call:\n",
      "lm(formula = model_pa, data = sample)\n",
      "\n",
      "Residuals:\n",
      "      Min        1Q    Median        3Q       Max \n",
      "-0.034198 -0.010207  0.001502  0.008715  0.036385 \n",
      "\n",
      "Coefficients:\n",
      "            Estimate Std. Error t value Pr(>|t|)    \n",
      "(Intercept)  0.24293    0.49128   0.494  0.62414    \n",
      "c_l1         0.59308    0.13282   4.465 8.38e-05 ***\n",
      "y            0.37497    0.11889   3.154  0.00336 ** \n",
      "D2           0.06148    0.01358   4.528 6.97e-05 ***\n",
      "D3           0.07666    0.01253   6.119 6.05e-07 ***\n",
      "D4           0.09831    0.01071   9.176 1.01e-10 ***\n",
      "---\n",
      "Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1\n",
      "\n",
      "Residual standard error: 0.01663 on 34 degrees of freedom\n",
      "Multiple R-squared:  0.9518,\tAdjusted R-squared:  0.9447 \n",
      "F-statistic: 134.3 on 5 and 34 DF,  p-value: < 2.2e-16\n",
      "\n",
      "[1] \"Sum of squared residuals: 0.00940082823730798\"\n",
      "\n",
      "\tDurbin-Watson test\n",
      "\n",
      "data:  fit\n",
      "DW = 2.1008, p-value = 0.6056\n",
      "alternative hypothesis: true autocorrelation is greater than 0\n",
      "\n",
      "\n",
      "\tRESET test\n",
      "\n",
      "data:  fit\n",
      "RESET = 0.043059, df1 = 1, df2 = 33, p-value = 0.8369\n",
      "\n",
      "\n",
      "\tJarque Bera Test\n",
      "\n",
      "data:  resid(fit)\n",
      "X-squared = 0.31405, df = 2, p-value = 0.8547\n",
      "\n",
      "\n",
      "\tBreusch-Godfrey test for serial correlation of order up to 4\n",
      "\n",
      "data:  fit\n",
      "LM test = 6.2062, df = 4, p-value = 0.1843\n",
      "\n",
      "\n",
      "\tstudentized Breusch-Pagan test\n",
      "\n",
      "data:  fit\n",
      "BP = 0.5423, df = 5, p-value = 0.9905\n",
      "\n"
     ]
    }
   ],
   "source": [
    "statistics(fit1247_pa)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "No problems visible in diagnostic statistics for this model. However, this model has markedly higher residual sum of squares."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If we will test linear restrictions for this model, we will see that this model has to be rejected against the general model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<thead><tr><th scope=col>Res.Df</th><th scope=col>RSS</th><th scope=col>Df</th><th scope=col>Sum of Sq</th><th scope=col>F</th><th scope=col>Pr(&gt;F)</th></tr></thead>\n",
       "<tbody>\n",
       "\t<tr><td>34         </td><td>0.009400828</td><td>NA         </td><td>         NA</td><td>      NA   </td><td>        NA </td></tr>\n",
       "\t<tr><td>31         </td><td>0.006587997</td><td> 3         </td><td>0.002812831</td><td>4.411952   </td><td>0.01073086 </td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "\\begin{tabular}{r|llllll}\n",
       " Res.Df & RSS & Df & Sum of Sq & F & Pr(>F)\\\\\n",
       "\\hline\n",
       "\t 34          & 0.009400828 & NA          &          NA &       NA    &         NA \\\\\n",
       "\t 31          & 0.006587997 &  3          & 0.002812831 & 4.411952    & 0.01073086 \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "Res.Df | RSS | Df | Sum of Sq | F | Pr(>F) | \n",
       "|---|---|\n",
       "| 34          | 0.009400828 | NA          |          NA |       NA    |         NA  | \n",
       "| 31          | 0.006587997 |  3          | 0.002812831 | 4.411952    | 0.01073086  | \n",
       "\n",
       "\n"
      ],
      "text/plain": [
       "  Res.Df RSS         Df Sum of Sq   F        Pr(>F)    \n",
       "1 34     0.009400828 NA          NA       NA         NA\n",
       "2 31     0.006587997  3 0.002812831 4.411952 0.01073086"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "car::linearHypothesis(fit1247,c(\"c_l2 = 0\",\"y_l1 = 0\", \"y_l2 = 0\"),test=\"F\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### First differences"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Models above have issue that both $c_t$ and $y_t$ have an upward trend over time. In attemp to avoid problems such as spurious correlation we will use first differences of our variables:\n",
    "\n",
    "$$ \\Delta c_t = c_t - c_{t-1} = ln(C)_t - ln(C)_{t-1}$$\n",
    "$$ \\Delta y_t = y_t - y_{t-1} = ln(Y)_t - ln(Y)_{t-1}$$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Call:\n",
      "lm(formula = model_diff, data = sample_diff)\n",
      "\n",
      "Residuals:\n",
      "      Min        1Q    Median        3Q       Max \n",
      "-0.030886 -0.006200 -0.002503  0.006769  0.043959 \n",
      "\n",
      "Coefficients:\n",
      "             Estimate Std. Error t value Pr(>|t|)    \n",
      "(Intercept) -0.058465   0.007389  -7.913 4.00e-09 ***\n",
      "c_d2        -0.464370   0.131332  -3.536 0.001229 ** \n",
      "y_d1         0.239245   0.114635   2.087 0.044690 *  \n",
      "y_d2         0.475831   0.117325   4.056 0.000287 ***\n",
      "D2           0.046497   0.017043   2.728 0.010126 *  \n",
      "D3           0.090621   0.007690  11.784 2.28e-13 ***\n",
      "D4           0.109025   0.007386  14.762 4.28e-16 ***\n",
      "---\n",
      "Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1\n",
      "\n",
      "Residual standard error: 0.01419 on 33 degrees of freedom\n",
      "Multiple R-squared:  0.9367,\tAdjusted R-squared:  0.9251 \n",
      "F-statistic: 81.33 on 6 and 33 DF,  p-value: < 2.2e-16\n",
      "\n",
      "[1] \"Sum of squared residuals: 0.00664339490039673\"\n",
      "\n",
      "\tDurbin-Watson test\n",
      "\n",
      "data:  fit\n",
      "DW = 1.9347, p-value = 0.5308\n",
      "alternative hypothesis: true autocorrelation is greater than 0\n",
      "\n",
      "\n",
      "\tRESET test\n",
      "\n",
      "data:  fit\n",
      "RESET = 0.801, df1 = 1, df2 = 32, p-value = 0.3775\n",
      "\n",
      "\n",
      "\tJarque Bera Test\n",
      "\n",
      "data:  resid(fit)\n",
      "X-squared = 10.3, df = 2, p-value = 0.0058\n",
      "\n",
      "\n",
      "\tBreusch-Godfrey test for serial correlation of order up to 4\n",
      "\n",
      "data:  fit\n",
      "LM test = 3.2338, df = 4, p-value = 0.5195\n",
      "\n",
      "\n",
      "\tstudentized Breusch-Pagan test\n",
      "\n",
      "data:  fit\n",
      "BP = 5.6341, df = 6, p-value = 0.4654\n",
      "\n"
     ]
    }
   ],
   "source": [
    "sample_diff <- sample %>%\n",
    "  mutate(c_d1 = c - c_l1) %>%\n",
    "  mutate(c_d2 = c_l1 - c_l2) %>%\n",
    "  mutate(y_d1 = y - y_l1) %>%\n",
    "  mutate(y_d2 = y_l1 - y_l2)\n",
    "\n",
    "model_diff <- c_d1 ~ c_d2 + y_d1 + y_d2 + D2 + D3 + D4\n",
    "fit1247_diff <- lm(model_diff, data = sample_diff)\n",
    "\n",
    "statistics(fit1247_diff)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First differences model has coefficients for all explanatory variables significantly different from zero. Diagnostic statistics are similar to the general model (fit1247), the only issue being normality test."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<thead><tr><th scope=col>Res.Df</th><th scope=col>RSS</th><th scope=col>Df</th><th scope=col>Sum of Sq</th><th scope=col>F</th><th scope=col>Pr(&gt;F)</th></tr></thead>\n",
       "<tbody>\n",
       "\t<tr><td>33          </td><td>0.006643395 </td><td>NA          </td><td>          NA</td><td>       NA   </td><td>       NA   </td></tr>\n",
       "\t<tr><td>31          </td><td>0.006587997 </td><td> 2          </td><td>5.539796e-05</td><td>0.1303383   </td><td>0.8782769   </td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "\\begin{tabular}{r|llllll}\n",
       " Res.Df & RSS & Df & Sum of Sq & F & Pr(>F)\\\\\n",
       "\\hline\n",
       "\t 33           & 0.006643395  & NA           &           NA &        NA    &        NA   \\\\\n",
       "\t 31           & 0.006587997  &  2           & 5.539796e-05 & 0.1303383    & 0.8782769   \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "Res.Df | RSS | Df | Sum of Sq | F | Pr(>F) | \n",
       "|---|---|\n",
       "| 33           | 0.006643395  | NA           |           NA |        NA    |        NA    | \n",
       "| 31           | 0.006587997  |  2           | 5.539796e-05 | 0.1303383    | 0.8782769    | \n",
       "\n",
       "\n"
      ],
      "text/plain": [
       "  Res.Df RSS         Df Sum of Sq    F         Pr(>F)   \n",
       "1 33     0.006643395 NA           NA        NA        NA\n",
       "2 31     0.006587997  2 5.539796e-05 0.1303383 0.8782769"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "car::linearHypothesis(fit1247,c(\"c_l1 + c_l2 = 1\",\"y + y_l1 + y_l2 = 0\"),test=\"F\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ECM model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Call:\n",
      "lm(formula = model_ecm, data = sample_diff)\n",
      "\n",
      "Residuals:\n",
      "      Min        1Q    Median        3Q       Max \n",
      "-0.029710 -0.005864 -0.001334  0.005772  0.042917 \n",
      "\n",
      "Coefficients:\n",
      "             Estimate Std. Error t value Pr(>|t|)    \n",
      "(Intercept)  0.014305   0.448594   0.032  0.97477    \n",
      "c_d2        -0.416320   0.165454  -2.516  0.01725 *  \n",
      "y_d1         0.272000   0.134187   2.027  0.05133 .  \n",
      "y_d2         0.446402   0.134436   3.321  0.00231 ** \n",
      "c_l1        -0.085062   0.167994  -0.506  0.61620    \n",
      "y_l1         0.077625   0.153065   0.507  0.61565    \n",
      "D2           0.045384   0.017653   2.571  0.01516 *  \n",
      "D3           0.086441   0.011423   7.568 1.57e-08 ***\n",
      "D4           0.105978   0.009664  10.967 3.38e-12 ***\n",
      "---\n",
      "Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1\n",
      "\n",
      "Residual standard error: 0.01458 on 31 degrees of freedom\n",
      "Multiple R-squared:  0.9372,\tAdjusted R-squared:  0.921 \n",
      "F-statistic: 57.81 on 8 and 31 DF,  p-value: < 2.2e-16\n",
      "\n",
      "[1] \"Sum of squared residuals: 0.00658799694266834\"\n",
      "\n",
      "\tDurbin-Watson test\n",
      "\n",
      "data:  fit\n",
      "DW = 1.8641, p-value = 0.3285\n",
      "alternative hypothesis: true autocorrelation is greater than 0\n",
      "\n",
      "\n",
      "\tRESET test\n",
      "\n",
      "data:  fit\n",
      "RESET = 0.56035, df1 = 1, df2 = 30, p-value = 0.4599\n",
      "\n",
      "\n",
      "\tJarque Bera Test\n",
      "\n",
      "data:  resid(fit)\n",
      "X-squared = 8.3655, df = 2, p-value = 0.01526\n",
      "\n",
      "\n",
      "\tBreusch-Godfrey test for serial correlation of order up to 4\n",
      "\n",
      "data:  fit\n",
      "LM test = 2.9251, df = 4, p-value = 0.5704\n",
      "\n",
      "\n",
      "\tstudentized Breusch-Pagan test\n",
      "\n",
      "data:  fit\n",
      "BP = 6.3187, df = 8, p-value = 0.6116\n",
      "\n"
     ]
    }
   ],
   "source": [
    "model_ecm <- c_d1 ~ c_d2 + y_d1 + y_d2 + c_l1 + y_l1 + D2 + D3 + D4\n",
    "fit1247_ecm <- lm(model_ecm, data = sample_diff)\n",
    "\n",
    "statistics(fit1247_ecm)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Model is simmilar to first differences model as newly added level variables coefficients are not significantly different from zero."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Call:\n",
      "lm(formula = model_ecm_fo, data = sample_diff)\n",
      "\n",
      "Residuals:\n",
      "      Min        1Q    Median        3Q       Max \n",
      "-0.033477 -0.009674  0.001033  0.008547  0.036918 \n",
      "\n",
      "Coefficients:\n",
      "            Estimate Std. Error t value Pr(>|t|)    \n",
      "(Intercept)  0.24198    0.49807   0.486 0.630303    \n",
      "y_d1         0.35147    0.14618   2.404 0.021975 *  \n",
      "c_l1        -0.42726    0.15249  -2.802 0.008435 ** \n",
      "y_l1         0.39522    0.14001   2.823 0.008004 ** \n",
      "D2           0.06080    0.01397   4.353 0.000122 ***\n",
      "D3           0.07622    0.01280   5.956 1.10e-06 ***\n",
      "D4           0.09835    0.01086   9.054 1.84e-10 ***\n",
      "---\n",
      "Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1\n",
      "\n",
      "Residual standard error: 0.01686 on 33 degrees of freedom\n",
      "Multiple R-squared:  0.9106,\tAdjusted R-squared:  0.8943 \n",
      "F-statistic: 56.01 on 6 and 33 DF,  p-value: 6.777e-16\n",
      "\n",
      "[1] \"Sum of squared residuals: 0.00937787024321339\"\n",
      "\n",
      "\tDurbin-Watson test\n",
      "\n",
      "data:  fit\n",
      "DW = 2.0086, p-value = 0.4928\n",
      "alternative hypothesis: true autocorrelation is greater than 0\n",
      "\n",
      "\n",
      "\tRESET test\n",
      "\n",
      "data:  fit\n",
      "RESET = 0.059217, df1 = 1, df2 = 32, p-value = 0.8093\n",
      "\n",
      "\n",
      "\tJarque Bera Test\n",
      "\n",
      "data:  resid(fit)\n",
      "X-squared = 0.26289, df = 2, p-value = 0.8768\n",
      "\n",
      "\n",
      "\tBreusch-Godfrey test for serial correlation of order up to 4\n",
      "\n",
      "data:  fit\n",
      "LM test = 7.1523, df = 4, p-value = 0.1281\n",
      "\n",
      "\n",
      "\tstudentized Breusch-Pagan test\n",
      "\n",
      "data:  fit\n",
      "BP = 1.1508, df = 6, p-value = 0.9792\n",
      "\n"
     ]
    }
   ],
   "source": [
    "model_ecm_fo <- c_d1 ~ y_d1 +  c_l1 + y_l1 + D2 + D3 + D4\n",
    "fit1247_ecm_fo <- lm(model_ecm_fo, data = sample_diff)\n",
    "\n",
    "statistics(fit1247_ecm_fo)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "With this simpler model coeficients for the level are significant. However, again error level has risen."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table>\n",
       "<thead><tr><th scope=col>Res.Df</th><th scope=col>RSS</th><th scope=col>Df</th><th scope=col>Sum of Sq</th><th scope=col>F</th><th scope=col>Pr(&gt;F)</th></tr></thead>\n",
       "<tbody>\n",
       "\t<tr><td>33         </td><td>0.009377870</td><td>NA         </td><td>         NA</td><td>      NA   </td><td>        NA </td></tr>\n",
       "\t<tr><td>31         </td><td>0.006587997</td><td> 2         </td><td>0.002789873</td><td>6.563913   </td><td>0.00419819 </td></tr>\n",
       "</tbody>\n",
       "</table>\n"
      ],
      "text/latex": [
       "\\begin{tabular}{r|llllll}\n",
       " Res.Df & RSS & Df & Sum of Sq & F & Pr(>F)\\\\\n",
       "\\hline\n",
       "\t 33          & 0.009377870 & NA          &          NA &       NA    &         NA \\\\\n",
       "\t 31          & 0.006587997 &  2          & 0.002789873 & 6.563913    & 0.00419819 \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/markdown": [
       "\n",
       "Res.Df | RSS | Df | Sum of Sq | F | Pr(>F) | \n",
       "|---|---|\n",
       "| 33          | 0.009377870 | NA          |          NA |       NA    |         NA  | \n",
       "| 31          | 0.006587997 |  2          | 0.002789873 | 6.563913    | 0.00419819  | \n",
       "\n",
       "\n"
      ],
      "text/plain": [
       "  Res.Df RSS         Df Sum of Sq   F        Pr(>F)    \n",
       "1 33     0.009377870 NA          NA       NA         NA\n",
       "2 31     0.006587997  2 0.002789873 6.563913 0.00419819"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "car::linearHypothesis(fit1247_ecm,c(\"y_d2 = 0\",\"c_d2 = 0\"),test=\"F\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "F-test indicates that this model has to be rejected in favor of fit1247_ecm_fo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "R",
   "language": "R",
   "name": "ir"
  },
  "language_info": {
   "codemirror_mode": "r",
   "file_extension": ".r",
   "mimetype": "text/x-r-source",
   "name": "R",
   "pygments_lexer": "r",
   "version": "3.5.0"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": false,
   "sideBar": false,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": false,
   "toc_window_display": false
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "state": {},
    "version_major": 2,
    "version_minor": 0
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}