{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This notebook continues on the topic of regression and covers: multiple predictors, categorial predictors, and interactions.\n",
    "\n",
    "# Multiple predictors\n",
    "\n",
    "Up to now, we have had one predictor for the dependent variable: To what extent does study time correlate with, and predict exam score? But often we have multiple predictors that we suspect to play a role. Hinton's example is: Both the student's intelligence and the time they spent studying sound like reasonable predictors for their exam scores (where intelligence is quantified as an IQ score). How to quantify this? The problem is that in order to get good predictions of the correlation, we need to know how study time influences exam score once intelligence is taken out of the picture. \n",
    "\n",
    "Here is how to work through this in Python. First, here is the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>participant</th>\n",
       "      <th>studytime</th>\n",
       "      <th>examscore</th>\n",
       "      <th>iq</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>40</td>\n",
       "      <td>58</td>\n",
       "      <td>118</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>43</td>\n",
       "      <td>73</td>\n",
       "      <td>128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>18</td>\n",
       "      <td>56</td>\n",
       "      <td>110</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>10</td>\n",
       "      <td>47</td>\n",
       "      <td>114</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>25</td>\n",
       "      <td>58</td>\n",
       "      <td>138</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6</td>\n",
       "      <td>33</td>\n",
       "      <td>54</td>\n",
       "      <td>120</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>7</td>\n",
       "      <td>27</td>\n",
       "      <td>45</td>\n",
       "      <td>106</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8</td>\n",
       "      <td>17</td>\n",
       "      <td>32</td>\n",
       "      <td>124</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>9</td>\n",
       "      <td>30</td>\n",
       "      <td>68</td>\n",
       "      <td>132</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>10</td>\n",
       "      <td>47</td>\n",
       "      <td>69</td>\n",
       "      <td>130</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   participant  studytime  examscore   iq\n",
       "0            1         40         58  118\n",
       "1            2         43         73  128\n",
       "2            3         18         56  110\n",
       "3            4         10         47  114\n",
       "4            5         25         58  138\n",
       "5            6         33         54  120\n",
       "6            7         27         45  106\n",
       "7            8         17         32  124\n",
       "8            9         30         68  132\n",
       "9           10         47         69  130"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "studytime_df = pd.DataFrame({\"participant\":[1,2,3,4,5,6,7,8,9,10],\n",
    "                            \"studytime\": [40,43,18,10,25,33,27,17,30,47],\n",
    "                            \"examscore\": [58,73,56,47,58,54,45,32,68,69],\n",
    "                            \"iq\":[118,128,110,114,138,120,106,124,132,130]})\n",
    "\n",
    "studytime_df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First step: Are the different predictors correlated with each other? And with the dependent variable? We already know that study time in this dataset correlates with exam score. So to what extent does intelligence correlate with study time, and with exam performance? "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correlation of study time and IQ:\n",
      "(0.37314321429686065, 0.28822074809763004)\n",
      "Correlation of IQ and exam performance:\n",
      "(0.4832859281320025, 0.1570552699191584)\n"
     ]
    }
   ],
   "source": [
    "from scipy import stats\n",
    "\n",
    "print(\"Correlation of study time and IQ:\")\n",
    "print(stats.pearsonr(studytime_df.studytime, studytime_df.iq))\n",
    "print(\"Correlation of IQ and exam performance:\")\n",
    "print(stats.pearsonr(studytime_df.iq, studytime_df.examscore))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The correlation is sizeable in both cases, though not statistically significant (as we have little data). \n",
    "\n",
    "We would like to determine the amount of variance in exam score that is predicted by study time after IQ has been taken out of the picture. To do that, we first take IQ out of the picture by predicting both study time and eam score from IQ and using the residuals, that is, the variance in study time and in exam score that is not explained by IQ."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.6652977419205109, 0.0357812018420546)"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import statsmodels.formula.api as smf\n",
    "\n",
    "# predicting study time from IQ\n",
    "ols_studytime = smf.ols(\"studytime ~ iq\", data= studytime_df).fit()\n",
    "# predicting exam score from IQ\n",
    "ols_examscore = smf.ols(\"examscore ~ iq\", data= studytime_df).fit()\n",
    "\n",
    "# computing the residuals\n",
    "residual_studytime = ols_studytime.resid\n",
    "residual_examscore = ols_examscore.resid\n",
    "\n",
    "# are they correlated?\n",
    "stats.pearsonr(residual_studytime, residual_examscore)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Yes, there is a strong correlation between the variance in study time not explained by IQ, and the variance in exam score not explained by IQ. \n",
    "\n",
    "The formula for multiple linear regression for two predictors has the form\n",
    "\n",
    "$Y = \\beta_0 + \\beta_1 x_1 + \\beta_2 x_2$\n",
    "\n",
    "and so on for more predictors.\n",
    "\n",
    "The linear regression implementation in ```statsmodels``` can deal with multiple predictors. Here is how to put multiple predictors into the \"formula\" format that ```ols()``` takes as input:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/scipy/stats/stats.py:1450: UserWarning: kurtosistest only valid for n>=20 ... continuing anyway, n=10\n",
      "  \"anyway, n=%i\" % int(n))\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>        <td>examscore</td>    <th>  R-squared:         </th> <td>   0.573</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.451</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   4.693</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Thu, 15 Apr 2021</td> <th>  Prob (F-statistic):</th>  <td>0.0510</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>11:33:05</td>     <th>  Log-Likelihood:    </th> <td> -34.616</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    10</td>      <th>  AIC:               </th> <td>   75.23</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>     7</td>      <th>  BIC:               </th> <td>   76.14</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>Intercept</th> <td>    0.2966</td> <td>   37.328</td> <td>    0.008</td> <td> 0.994</td> <td>  -87.969</td> <td>   88.562</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>studytime</th> <td>    0.6486</td> <td>    0.275</td> <td>    2.358</td> <td> 0.051</td> <td>   -0.002</td> <td>    1.299</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>iq</th>        <td>    0.3024</td> <td>    0.323</td> <td>    0.935</td> <td> 0.381</td> <td>   -0.462</td> <td>    1.067</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 1.671</td> <th>  Durbin-Watson:     </th> <td>   1.786</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.434</td> <th>  Jarque-Bera (JB):  </th> <td>   0.609</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td>-0.602</td> <th>  Prob(JB):          </th> <td>   0.737</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 2.880</td> <th>  Cond. No.          </th> <td>1.61e+03</td>\n",
       "</tr>\n",
       "</table><br/><br/>Warnings:<br/>[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.<br/>[2] The condition number is large, 1.61e+03. This might indicate that there are<br/>strong multicollinearity or other numerical problems."
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                            OLS Regression Results                            \n",
       "==============================================================================\n",
       "Dep. Variable:              examscore   R-squared:                       0.573\n",
       "Model:                            OLS   Adj. R-squared:                  0.451\n",
       "Method:                 Least Squares   F-statistic:                     4.693\n",
       "Date:                Thu, 15 Apr 2021   Prob (F-statistic):             0.0510\n",
       "Time:                        11:33:05   Log-Likelihood:                -34.616\n",
       "No. Observations:                  10   AIC:                             75.23\n",
       "Df Residuals:                       7   BIC:                             76.14\n",
       "Df Model:                           2                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "==============================================================================\n",
       "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "Intercept      0.2966     37.328      0.008      0.994     -87.969      88.562\n",
       "studytime      0.6486      0.275      2.358      0.051      -0.002       1.299\n",
       "iq             0.3024      0.323      0.935      0.381      -0.462       1.067\n",
       "==============================================================================\n",
       "Omnibus:                        1.671   Durbin-Watson:                   1.786\n",
       "Prob(Omnibus):                  0.434   Jarque-Bera (JB):                0.609\n",
       "Skew:                          -0.602   Prob(JB):                        0.737\n",
       "Kurtosis:                       2.880   Cond. No.                     1.61e+03\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "[2] The condition number is large, 1.61e+03. This might indicate that there are\n",
       "strong multicollinearity or other numerical problems.\n",
       "\"\"\""
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ols_twopredictors = smf.ols(\"examscore ~ studytime + iq\", data= studytime_df).fit()\n",
    "ols_twopredictors.summary()"
   ]
  },
  {
   "attachments": {
    "Screen%20Shot%202021-04-15%20at%206.07.25%20PM.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "What do we see here?\n",
    "\n",
    "* The coefficients for study time and IQ are both positive: higher study time and higher IQ both go with higher exam scores.\n",
    "* Looking at the t tests for both coefficients, we see however that the slope for IQ is not significantly different from zero -- this is not a reliable predictor for exam score.\n",
    "* The intercept is not interpretable when we have multiple predictors. \n",
    "* Residuals: Looking at the Omnibus and Jarque-Bera tests, the residuals again fail the test for normality (unsurprisingly, as this is a variant of the same dataset that failed normality of its residuals last time)\n",
    "* Amount of variance explained: We get an R-squared value of 0.573. Now that we have multiple predictors, we see that the adjusted R-squared value, which adjusts for random variations in the data, is now quite a bit lower than the non-adjusted R-squared, and is at 0.451: We explain about 45% of the variance in exam scores with the predictors that we have. \n",
    "* How good is the model overall? Can we trust that not all slopes are really zero? The F-test shows a p-value of 0.051, which shows that the model probably does have value, though it is a bit borderline. This is even though the correlation of study time and exam score is so strong that you can actually see it in a plot of the data. The reason the model is dubious is that we have so few data points.\n",
    "\n",
    "\n",
    "We now have a new problem, shown at the very bottom. There it says \"The condition number is large, 1.61e+03. This might indicate that there are strong multicollinearity or other numerical problems.\" In fact, if the collinearity is above 20, that indicates worrying level of collinearity.\n",
    "\n",
    "You can find the condition number in the last table, under **Cond. No.** You should always check the condition number to make sure you do not have an unduly high level of collinearity among your predictors. \n",
    "\n",
    "What should you do when you have a high level of collinearity among your predictors?\n",
    "* First, this is a problem when you care to inspect the coefficients, that is, when you are doing data analysis. When you only care about the prediction result, collinearity is not a problem, because in that case you do not care so much about what the coefficient is for each predictor.\n",
    "* If you are about the coefficients, then this is a problem. In that case, first think about your predictors: You have some predictors that tell pretty much the same story. Which ones are they? Can you just drop some of them? \n",
    "* If you decide you cannot and do not want to drop any of them, another option is to run principal component analysis over your predictors to obtain a new set of predictors that are mutually orthogonal. \n",
    "\n",
    "# ols output, annotated\n",
    "\n",
    "![Screen%20Shot%202021-04-15%20at%206.07.25%20PM.png](attachment:Screen%20Shot%202021-04-15%20at%206.07.25%20PM.png)\n",
    "\n",
    "# Another dataset\n",
    "\n",
    "**Try it for yourself:**\n",
    "\n",
    "On Canvas, you find the file ```lexdec2.csv```, an extension of the lexical decision data we used before. It contains additional predictors, including the participants' subjective frequency ratings of words. The column ```SubjFreq``` contains these subjective frequencies, averaged over subjects.\n",
    "\n",
    "Fit a linear regression model that uses both ```Frequency``` and ```SubjFreq``` as predictors for reaction time. \n",
    "\n",
    "* Do both coefficients have a slope that is significantly different from zero? \n",
    "* How much of the variance in reaction time do you explain with this model? (Please make sure to use the value that is adjusted for the number of predictors.) Is this better than the variance explained by the model that used only frequency as a predictor? \n",
    "* Does the model as a whole have value? Where do you see this?\n",
    "* Are the residuals approximately normally distributed? How do you determine this? "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>Word</th>\n",
       "      <th>Frequency</th>\n",
       "      <th>meanRT</th>\n",
       "      <th>SubjFreq</th>\n",
       "      <th>Class</th>\n",
       "      <th>Complex</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>owl</td>\n",
       "      <td>4.859812</td>\n",
       "      <td>6.3582</td>\n",
       "      <td>3.12</td>\n",
       "      <td>animal</td>\n",
       "      <td>simplex</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>mole</td>\n",
       "      <td>4.605170</td>\n",
       "      <td>6.4150</td>\n",
       "      <td>2.40</td>\n",
       "      <td>animal</td>\n",
       "      <td>simplex</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>cherry</td>\n",
       "      <td>4.997212</td>\n",
       "      <td>6.3426</td>\n",
       "      <td>3.88</td>\n",
       "      <td>plant</td>\n",
       "      <td>simplex</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>pear</td>\n",
       "      <td>4.727388</td>\n",
       "      <td>6.3353</td>\n",
       "      <td>4.52</td>\n",
       "      <td>plant</td>\n",
       "      <td>simplex</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>dog</td>\n",
       "      <td>7.667626</td>\n",
       "      <td>6.2956</td>\n",
       "      <td>6.04</td>\n",
       "      <td>animal</td>\n",
       "      <td>simplex</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Word  Frequency  meanRT  SubjFreq   Class  Complex\n",
       "0     owl   4.859812  6.3582      3.12  animal  simplex\n",
       "1    mole   4.605170  6.4150      2.40  animal  simplex\n",
       "2  cherry   4.997212  6.3426      3.88   plant  simplex\n",
       "3    pear   4.727388  6.3353      4.52   plant  simplex\n",
       "4     dog   7.667626  6.2956      6.04  animal  simplex"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lexdec2_df = pd.read_csv(\"lexdec2.csv\")\n",
    "lexdec2_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Categorial predictors\n",
    "\n",
    "A categorial variable is a variable that takes one one of several distinct categories: yes versus no, blue versus yellow versus green, Texas versus Colorado versus New Mexico. \n",
    "\n",
    "Categorial variables are a problem because we somehow need to turn them into numbers in order to use them as predictors in a regression. But if we just encode, say, the state of origin as a number between 1 and 50, that indicates that states 1 and 2 are somehow closer together  than states 1 and 49. But that makes no sense for categories.\n",
    "\n",
    "Here is an example to illustrate this: Say there have been four different package design for a new cereal, which were tested across a number of stores. Each store is randomly assigned a package design. Then we would like to know what the connection is between the package design and the number of items sold -- did some designs sell better than others? Here is a faulty encoding that just encodes blue as 1, red as 2, yellow as 3, and green as 4. We make up some sales numbers at random by drawing a number between 10 and 30 from a uniform distribution. First, here is how we draw 20 random integers between 10 and 30:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Here are 20 random numbers between 0 and 1 [0.92365323 0.50499755 0.81465005 0.91652947 0.78754762 0.0983289\n",
      " 0.58256167 0.43721868 0.85807323 0.71901738 0.31263082 0.92505232\n",
      " 0.565689   0.54368638 0.85456825 0.99118295 0.90244907 0.53640021\n",
      " 0.639402   0.90293479]\n",
      "Here are 20 random numbers between 10 and 30 [17.31127331 21.09814648 14.00019978 26.88181994 28.00125672 24.04358013\n",
      " 11.06756614 29.35417117 24.19600962 23.40675603 23.89773992 28.58714708\n",
      " 11.69962257 25.44360575 12.83877929 13.01877639 16.49801277 24.6725053\n",
      " 24.9628653  16.86961988]\n",
      "Here are 20 random integers between 10 and 30 [23.0, 16.0, 19.0, 18.0, 18.0, 12.0, 24.0, 27.0, 18.0, 15.0, 11.0, 11.0, 17.0, 21.0, 24.0, 27.0, 28.0, 13.0, 17.0, 18.0]\n"
     ]
    }
   ],
   "source": [
    "from scipy import stats\n",
    "\n",
    "# this draws 20 random numbers between 0 and 1\n",
    "print(\"Here are 20 random numbers between 0 and 1\", stats.uniform.rvs(size = 20))\n",
    "# this draws 20 random numbers between 10 and 30\n",
    "print(\"Here are 20 random numbers between 10 and 30\", stats.uniform.rvs(loc=10, scale = 20, size = 20))\n",
    "# and this draws 20 random numbers between 10 and 30, integers only\n",
    "print(\"Here are 20 random integers between 10 and 30\", \n",
    "     [round(val) for val in stats.uniform.rvs(loc = 10, scale = 20, size = 20)])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We use this to make up our fake data for 20 stores: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "random_sales = [round(val) for val in stats.uniform.rvs(loc = 10, scale = 20, size = 20)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we make a *broken* data frame in which we wrongly code colors as numbers. 5 stores get the yellow design, 5 the blue, 5 the green, and 5 the red:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "\n",
    "cereal_df_broken = pd.DataFrame( { \"design\" : [1,1,1,1,1,2,2,2,2,2,3,3,3,3,3,4,4,4,4,4],\n",
    "                            \"sales\":random_sales})\n",
    "\n",
    "# let's plot this\n",
    "%matplotlib inline\n",
    "cereal_df_broken.plot(x = \"design\", y = \"sales\", style = \"o\", xlim = (0.5, 4.5), legend = False);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So, what does linear regression make of this?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>          <td>sales</td>      <th>  R-squared:         </th> <td>   0.036</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>  -0.017</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>  0.6790</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Thu, 15 Apr 2021</td> <th>  Prob (F-statistic):</th>  <td> 0.421</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>11:33:05</td>     <th>  Log-Likelihood:    </th> <td> -62.982</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    20</td>      <th>  AIC:               </th> <td>   130.0</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    18</td>      <th>  BIC:               </th> <td>   132.0</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>Intercept</th> <td>   17.9000</td> <td>    3.257</td> <td>    5.496</td> <td> 0.000</td> <td>   11.057</td> <td>   24.743</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>design</th>    <td>    0.9800</td> <td>    1.189</td> <td>    0.824</td> <td> 0.421</td> <td>   -1.519</td> <td>    3.479</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 0.016</td> <th>  Durbin-Watson:     </th> <td>   2.100</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.992</td> <th>  Jarque-Bera (JB):  </th> <td>   0.223</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.001</td> <th>  Prob(JB):          </th> <td>   0.895</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 2.483</td> <th>  Cond. No.          </th> <td>    7.47</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:                  sales   R-squared:                       0.036\n",
       "Model:                            OLS   Adj. R-squared:                 -0.017\n",
       "Method:                 Least Squares   F-statistic:                    0.6790\n",
       "Date:                Thu, 15 Apr 2021   Prob (F-statistic):              0.421\n",
       "Time:                        11:33:05   Log-Likelihood:                -62.982\n",
       "No. Observations:                  20   AIC:                             130.0\n",
       "Df Residuals:                      18   BIC:                             132.0\n",
       "Df Model:                           1                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "==============================================================================\n",
       "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------\n",
       "Intercept     17.9000      3.257      5.496      0.000      11.057      24.743\n",
       "design         0.9800      1.189      0.824      0.421      -1.519       3.479\n",
       "==============================================================================\n",
       "Omnibus:                        0.016   Durbin-Watson:                   2.100\n",
       "Prob(Omnibus):                  0.992   Jarque-Bera (JB):                0.223\n",
       "Skew:                           0.001   Prob(JB):                        0.895\n",
       "Kurtosis:                       2.483   Cond. No.                         7.47\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import statsmodels.formula.api as smf\n",
    "\n",
    "smf.ols(\"sales ~ design\", data= cereal_df_broken).fit().summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This hallucinates some relation about how sales change linearly as the design number rises by one unit -- which makes no sense at all. \n",
    "\n",
    "Instead we need to let ```ols()``` know that design is a categorial variable:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>          <td>sales</td>      <th>  R-squared:         </th> <td>   0.281</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.146</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   2.080</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Thu, 15 Apr 2021</td> <th>  Prob (F-statistic):</th>  <td> 0.143</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>11:33:05</td>     <th>  Log-Likelihood:    </th> <td> -60.059</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    20</td>      <th>  AIC:               </th> <td>   128.1</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    16</td>      <th>  BIC:               </th> <td>   132.1</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Model:</th>              <td>     3</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>Intercept</th>        <td>   16.0000</td> <td>    2.437</td> <td>    6.565</td> <td> 0.000</td> <td>   10.833</td> <td>   21.167</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>design[T.green]</th>  <td>    8.6000</td> <td>    3.447</td> <td>    2.495</td> <td> 0.024</td> <td>    1.293</td> <td>   15.907</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>design[T.red]</th>    <td>    4.6000</td> <td>    3.447</td> <td>    1.335</td> <td> 0.201</td> <td>   -2.707</td> <td>   11.907</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>design[T.yellow]</th> <td>    4.2000</td> <td>    3.447</td> <td>    1.219</td> <td> 0.241</td> <td>   -3.107</td> <td>   11.507</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 0.277</td> <th>  Durbin-Watson:     </th> <td>   2.588</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.871</td> <th>  Jarque-Bera (JB):  </th> <td>   0.444</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.188</td> <th>  Prob(JB):          </th> <td>   0.801</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 2.374</td> <th>  Cond. No.          </th> <td>    4.79</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:                  sales   R-squared:                       0.281\n",
       "Model:                            OLS   Adj. R-squared:                  0.146\n",
       "Method:                 Least Squares   F-statistic:                     2.080\n",
       "Date:                Thu, 15 Apr 2021   Prob (F-statistic):              0.143\n",
       "Time:                        11:33:05   Log-Likelihood:                -60.059\n",
       "No. Observations:                  20   AIC:                             128.1\n",
       "Df Residuals:                      16   BIC:                             132.1\n",
       "Df Model:                           3                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "====================================================================================\n",
       "                       coef    std err          t      P>|t|      [0.025      0.975]\n",
       "------------------------------------------------------------------------------------\n",
       "Intercept           16.0000      2.437      6.565      0.000      10.833      21.167\n",
       "design[T.green]      8.6000      3.447      2.495      0.024       1.293      15.907\n",
       "design[T.red]        4.6000      3.447      1.335      0.201      -2.707      11.907\n",
       "design[T.yellow]     4.2000      3.447      1.219      0.241      -3.107      11.507\n",
       "==============================================================================\n",
       "Omnibus:                        0.277   Durbin-Watson:                   2.588\n",
       "Prob(Omnibus):                  0.871   Jarque-Bera (JB):                0.444\n",
       "Skew:                           0.188   Prob(JB):                        0.801\n",
       "Kurtosis:                       2.374   Cond. No.                         4.79\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cereal_df = pd.DataFrame( { \"design\" : [\"yellow\",\"yellow\", \"yellow\", \"yellow\", \"yellow\",\n",
    "                                        \"blue\", \"blue\", \"blue\", \"blue\", \"blue\",\n",
    "                                        \"green\", \"green\", \"green\", \"green\", \"green\", \n",
    "                                        \"red\", \"red\",\"red\",\"red\",\"red\"],\n",
    "                           \"sales\":random_sales})\n",
    "\n",
    "smf.ols(\"sales ~ design\", data= cereal_df).fit().summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here is how to read this: One category's mean becomes the intercept. Here it is blue. The \"slope\" on all other categories is the difference from the mean value for blue packages. As can be seen from the probabilities associated with the t-tests for all coefficients, there is no significant difference between categories in this case -- which makes sense, since we randomly drew the sales values for all four categories from a uniform distribution. \n",
    "\n",
    "# An actual dataset with categorial predictors: lexical decision times, again\n",
    "\n",
    "\n",
    "We again use the file ```lexdec2.csv``` with data on lexical decision times. There are two additional predictors: ```Complex``` says whether the word is morphologically complex. It has two categorial values, \"complex\" and \"simplex\". ```Class``` contains the category that the word came from. This whole dataset only has animals and plants, and ```Class``` encodes whether the word describes an animal or plant.\n",
    "\n",
    "Choose one of the two categorial variables, and use it as a predictor of reaction time. First use it on its own, then together with Frequency.\n",
    "\n",
    "* What list of coefficients do you get? what are they saying? Is each coefficient truly different from zero?\n",
    "* How much of the variance in reaction time do you explain with this model? (Please make sure to use the value that is adjusted for the number of predictors.) Is this better than the variance explained by the model that used only frequency as a predictor? \n",
    "* Does the model as a whole have value? Where do you see this?\n",
    "* Are the residuals approximately normally distributed? How do you determine this? \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>Word</th>\n",
       "      <th>Frequency</th>\n",
       "      <th>meanRT</th>\n",
       "      <th>SubjFreq</th>\n",
       "      <th>Class</th>\n",
       "      <th>Complex</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>owl</td>\n",
       "      <td>4.859812</td>\n",
       "      <td>6.3582</td>\n",
       "      <td>3.12</td>\n",
       "      <td>animal</td>\n",
       "      <td>simplex</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>mole</td>\n",
       "      <td>4.605170</td>\n",
       "      <td>6.4150</td>\n",
       "      <td>2.40</td>\n",
       "      <td>animal</td>\n",
       "      <td>simplex</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>cherry</td>\n",
       "      <td>4.997212</td>\n",
       "      <td>6.3426</td>\n",
       "      <td>3.88</td>\n",
       "      <td>plant</td>\n",
       "      <td>simplex</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>pear</td>\n",
       "      <td>4.727388</td>\n",
       "      <td>6.3353</td>\n",
       "      <td>4.52</td>\n",
       "      <td>plant</td>\n",
       "      <td>simplex</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>dog</td>\n",
       "      <td>7.667626</td>\n",
       "      <td>6.2956</td>\n",
       "      <td>6.04</td>\n",
       "      <td>animal</td>\n",
       "      <td>simplex</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Word  Frequency  meanRT  SubjFreq   Class  Complex\n",
       "0     owl   4.859812  6.3582      3.12  animal  simplex\n",
       "1    mole   4.605170  6.4150      2.40  animal  simplex\n",
       "2  cherry   4.997212  6.3426      3.88   plant  simplex\n",
       "3    pear   4.727388  6.3353      4.52   plant  simplex\n",
       "4     dog   7.667626  6.2956      6.04  animal  simplex"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "lexdec_df = pd.read_csv(\"lexdec2.csv\")\n",
    "lexdec_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5     blackberry\n",
       "6     strawberry\n",
       "11      reindeer\n",
       "12     blueberry\n",
       "16      eggplant\n",
       "18        peanut\n",
       "33      beetroot\n",
       "37     pineapple\n",
       "68    woodpecker\n",
       "73     butterfly\n",
       "Name: Word, dtype: object"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lexdec_df[lexdec_df.Complex == \"complex\"].Word"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>         <td>meanRT</td>      <th>  R-squared:         </th> <td>   0.013</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>  -0.000</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>  0.9937</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Thu, 15 Apr 2021</td> <th>  Prob (F-statistic):</th>  <td> 0.322</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>12:07:44</td>     <th>  Log-Likelihood:    </th> <td>  91.141</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    79</td>      <th>  AIC:               </th> <td>  -178.3</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    77</td>      <th>  BIC:               </th> <td>  -173.5</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>Intercept</th>          <td>    6.4016</td> <td>    0.024</td> <td>  261.817</td> <td> 0.000</td> <td>    6.353</td> <td>    6.450</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Complex[T.simplex]</th> <td>   -0.0261</td> <td>    0.026</td> <td>   -0.997</td> <td> 0.322</td> <td>   -0.078</td> <td>    0.026</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>11.242</td> <th>  Durbin-Watson:     </th> <td>   2.105</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.004</td> <th>  Jarque-Bera (JB):  </th> <td>  11.433</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.869</td> <th>  Prob(JB):          </th> <td> 0.00329</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 3.672</td> <th>  Cond. No.          </th> <td>    5.45</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:                 meanRT   R-squared:                       0.013\n",
       "Model:                            OLS   Adj. R-squared:                 -0.000\n",
       "Method:                 Least Squares   F-statistic:                    0.9937\n",
       "Date:                Thu, 15 Apr 2021   Prob (F-statistic):              0.322\n",
       "Time:                        12:07:44   Log-Likelihood:                 91.141\n",
       "No. Observations:                  79   AIC:                            -178.3\n",
       "Df Residuals:                      77   BIC:                            -173.5\n",
       "Df Model:                           1                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "======================================================================================\n",
       "                         coef    std err          t      P>|t|      [0.025      0.975]\n",
       "--------------------------------------------------------------------------------------\n",
       "Intercept              6.4016      0.024    261.817      0.000       6.353       6.450\n",
       "Complex[T.simplex]    -0.0261      0.026     -0.997      0.322      -0.078       0.026\n",
       "==============================================================================\n",
       "Omnibus:                       11.242   Durbin-Watson:                   2.105\n",
       "Prob(Omnibus):                  0.004   Jarque-Bera (JB):               11.433\n",
       "Skew:                           0.869   Prob(JB):                      0.00329\n",
       "Kurtosis:                       3.672   Cond. No.                         5.45\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ols_obj = smf.ols(\"meanRT ~ Complex\", data = lexdec_df).fit()\n",
    "ols_obj.summary()"
   ]
  },
  {
   "attachments": {
    "Screen%20Shot%202021-04-15%20at%206.09.51%20PM.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# ols output, annotated\n",
    "\n",
    "![Screen%20Shot%202021-04-15%20at%206.09.51%20PM.png](attachment:Screen%20Shot%202021-04-15%20at%206.09.51%20PM.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>         <td>meanRT</td>      <th>  R-squared:         </th> <td>   0.003</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>  -0.010</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>  0.2031</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Thu, 15 Apr 2021</td> <th>  Prob (F-statistic):</th>  <td> 0.654</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>12:15:31</td>     <th>  Log-Likelihood:    </th> <td>  90.738</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    79</td>      <th>  AIC:               </th> <td>  -177.5</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    77</td>      <th>  BIC:               </th> <td>  -172.7</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>Intercept</th>      <td>    6.3824</td> <td>    0.012</td> <td>  544.756</td> <td> 0.000</td> <td>    6.359</td> <td>    6.406</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Class[T.plant]</th> <td>   -0.0079</td> <td>    0.018</td> <td>   -0.451</td> <td> 0.654</td> <td>   -0.043</td> <td>    0.027</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td> 8.707</td> <th>  Durbin-Watson:     </th> <td>   2.175</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.013</td> <th>  Jarque-Bera (JB):  </th> <td>   8.507</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.786</td> <th>  Prob(JB):          </th> <td>  0.0142</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 3.338</td> <th>  Cond. No.          </th> <td>    2.51</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:                 meanRT   R-squared:                       0.003\n",
       "Model:                            OLS   Adj. R-squared:                 -0.010\n",
       "Method:                 Least Squares   F-statistic:                    0.2031\n",
       "Date:                Thu, 15 Apr 2021   Prob (F-statistic):              0.654\n",
       "Time:                        12:15:31   Log-Likelihood:                 90.738\n",
       "No. Observations:                  79   AIC:                            -177.5\n",
       "Df Residuals:                      77   BIC:                            -172.7\n",
       "Df Model:                           1                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "==================================================================================\n",
       "                     coef    std err          t      P>|t|      [0.025      0.975]\n",
       "----------------------------------------------------------------------------------\n",
       "Intercept          6.3824      0.012    544.756      0.000       6.359       6.406\n",
       "Class[T.plant]    -0.0079      0.018     -0.451      0.654      -0.043       0.027\n",
       "==============================================================================\n",
       "Omnibus:                        8.707   Durbin-Watson:                   2.175\n",
       "Prob(Omnibus):                  0.013   Jarque-Bera (JB):                8.507\n",
       "Skew:                           0.786   Prob(JB):                       0.0142\n",
       "Kurtosis:                       3.338   Cond. No.                         2.51\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# now, animals versus plants:\n",
    "ols_obj = smf.ols(\"meanRT ~ Class\", data = lexdec_df).fit()\n",
    "ols_obj.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>OLS Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>         <td>meanRT</td>      <th>  R-squared:         </th> <td>   0.467</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.453</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   33.32</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>             <td>Thu, 15 Apr 2021</td> <th>  Prob (F-statistic):</th> <td>4.08e-11</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                 <td>12:17:57</td>     <th>  Log-Likelihood:    </th> <td>  115.50</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Observations:</th>      <td>    79</td>      <th>  AIC:               </th> <td>  -225.0</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Df Residuals:</th>          <td>    76</td>      <th>  BIC:               </th> <td>  -217.9</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>Intercept</th>      <td>    6.6061</td> <td>    0.029</td> <td>  229.344</td> <td> 0.000</td> <td>    6.549</td> <td>    6.663</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Class[T.plant]</th> <td>   -0.0454</td> <td>    0.014</td> <td>   -3.302</td> <td> 0.001</td> <td>   -0.073</td> <td>   -0.018</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Frequency</th>      <td>   -0.0436</td> <td>    0.005</td> <td>   -8.140</td> <td> 0.000</td> <td>   -0.054</td> <td>   -0.033</td>\n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "  <th>Omnibus:</th>       <td>12.173</td> <th>  Durbin-Watson:     </th> <td>   2.272</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Prob(Omnibus):</th> <td> 0.002</td> <th>  Jarque-Bera (JB):  </th> <td>  16.432</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Skew:</th>          <td> 0.664</td> <th>  Prob(JB):          </th> <td>0.000270</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Kurtosis:</th>      <td> 4.797</td> <th>  Cond. No.          </th> <td>    23.6</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:                 meanRT   R-squared:                       0.467\n",
       "Model:                            OLS   Adj. R-squared:                  0.453\n",
       "Method:                 Least Squares   F-statistic:                     33.32\n",
       "Date:                Thu, 15 Apr 2021   Prob (F-statistic):           4.08e-11\n",
       "Time:                        12:17:57   Log-Likelihood:                 115.50\n",
       "No. Observations:                  79   AIC:                            -225.0\n",
       "Df Residuals:                      76   BIC:                            -217.9\n",
       "Df Model:                           2                                         \n",
       "Covariance Type:            nonrobust                                         \n",
       "==================================================================================\n",
       "                     coef    std err          t      P>|t|      [0.025      0.975]\n",
       "----------------------------------------------------------------------------------\n",
       "Intercept          6.6061      0.029    229.344      0.000       6.549       6.663\n",
       "Class[T.plant]    -0.0454      0.014     -3.302      0.001      -0.073      -0.018\n",
       "Frequency         -0.0436      0.005     -8.140      0.000      -0.054      -0.033\n",
       "==============================================================================\n",
       "Omnibus:                       12.173   Durbin-Watson:                   2.272\n",
       "Prob(Omnibus):                  0.002   Jarque-Bera (JB):               16.432\n",
       "Skew:                           0.664   Prob(JB):                     0.000270\n",
       "Kurtosis:                       4.797   Cond. No.                         23.6\n",
       "==============================================================================\n",
       "\n",
       "Warnings:\n",
       "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
       "\"\"\""
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# now, animals versus plants:\n",
    "ols_obj = smf.ols(\"meanRT ~ Class + Frequency\", data = lexdec_df).fit()\n",
    "ols_obj.summary()"
   ]
  },
  {
   "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}