{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Generalized Linear Models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true,
    "execution": {
     "iopub.execute_input": "2021-02-02T06:54:01.500221Z",
     "iopub.status.busy": "2021-02-02T06:54:01.499537Z",
     "iopub.status.idle": "2021-02-02T06:54:01.748687Z",
     "shell.execute_reply": "2021-02-02T06:54:01.748188Z"
    }
   },
   "outputs": [],
   "source": [
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-02-02T06:54:01.752971Z",
     "iopub.status.busy": "2021-02-02T06:54:01.752380Z",
     "iopub.status.idle": "2021-02-02T06:54:02.648733Z",
     "shell.execute_reply": "2021-02-02T06:54:02.649867Z"
    }
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import statsmodels.api as sm\n",
    "from scipy import stats\n",
    "from matplotlib import pyplot as plt\n",
    "\n",
    "plt.rc(\"figure\", figsize=(16,8))\n",
    "plt.rc(\"font\", size=14)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## GLM: Binomial response data\n",
    "\n",
    "### Load Star98 data\n",
    "\n",
    " In this example, we use the Star98 dataset which was taken with permission\n",
    " from Jeff Gill (2000) Generalized linear models: A unified approach. Codebook\n",
    " information can be obtained by typing: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-02-02T06:54:02.655277Z",
     "iopub.status.busy": "2021-02-02T06:54:02.653788Z",
     "iopub.status.idle": "2021-02-02T06:54:02.662855Z",
     "shell.execute_reply": "2021-02-02T06:54:02.664167Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "::\n",
      "\n",
      "    Number of Observations - 303 (counties in California).\n",
      "\n",
      "    Number of Variables - 13 and 8 interaction terms.\n",
      "\n",
      "    Definition of variables names::\n",
      "\n",
      "        NABOVE   - Total number of students above the national median for the\n",
      "                   math section.\n",
      "        NBELOW   - Total number of students below the national median for the\n",
      "                   math section.\n",
      "        LOWINC   - Percentage of low income students\n",
      "        PERASIAN - Percentage of Asian student\n",
      "        PERBLACK - Percentage of black students\n",
      "        PERHISP  - Percentage of Hispanic students\n",
      "        PERMINTE - Percentage of minority teachers\n",
      "        AVYRSEXP - Sum of teachers' years in educational service divided by the\n",
      "                number of teachers.\n",
      "        AVSALK   - Total salary budget including benefits divided by the number\n",
      "                   of full-time teachers (in thousands)\n",
      "        PERSPENK - Per-pupil spending (in thousands)\n",
      "        PTRATIO  - Pupil-teacher ratio.\n",
      "        PCTAF    - Percentage of students taking UC/CSU prep courses\n",
      "        PCTCHRT  - Percentage of charter schools\n",
      "        PCTYRRND - Percentage of year-round schools\n",
      "\n",
      "        The below variables are interaction terms of the variables defined\n",
      "        above.\n",
      "\n",
      "        PERMINTE_AVYRSEXP\n",
      "        PEMINTE_AVSAL\n",
      "        AVYRSEXP_AVSAL\n",
      "        PERSPEN_PTRATIO\n",
      "        PERSPEN_PCTAF\n",
      "        PTRATIO_PCTAF\n",
      "        PERMINTE_AVTRSEXP_AVSAL\n",
      "        PERSPEN_PTRATIO_PCTAF\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(sm.datasets.star98.NOTE)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Load the data and add a constant to the exogenous (independent) variables:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-02-02T06:54:02.672546Z",
     "iopub.status.busy": "2021-02-02T06:54:02.666966Z",
     "iopub.status.idle": "2021-02-02T06:54:02.689742Z",
     "shell.execute_reply": "2021-02-02T06:54:02.690579Z"
    }
   },
   "outputs": [],
   "source": [
    "data = sm.datasets.star98.load(as_pandas=False)\n",
    "data.exog = sm.add_constant(data.exog, prepend=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " The dependent variable is N by 2 (Success: NABOVE, Failure: NBELOW): "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-02-02T06:54:02.694327Z",
     "iopub.status.busy": "2021-02-02T06:54:02.693184Z",
     "iopub.status.idle": "2021-02-02T06:54:02.699862Z",
     "shell.execute_reply": "2021-02-02T06:54:02.700665Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[452. 355.]\n",
      " [144.  40.]\n",
      " [337. 234.]\n",
      " [395. 178.]\n",
      " [  8.  57.]]\n"
     ]
    }
   ],
   "source": [
    "print(data.endog[:5,:])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " The independent variables include all the other variables described above, as\n",
    " well as the interaction terms:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-02-02T06:54:02.704617Z",
     "iopub.status.busy": "2021-02-02T06:54:02.703466Z",
     "iopub.status.idle": "2021-02-02T06:54:02.710403Z",
     "shell.execute_reply": "2021-02-02T06:54:02.711191Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[3.43973000e+01 2.32993000e+01 1.42352800e+01 1.14111200e+01\n",
      "  1.59183700e+01 1.47064600e+01 5.91573200e+01 4.44520700e+00\n",
      "  2.17102500e+01 5.70327600e+01 0.00000000e+00 2.22222200e+01\n",
      "  2.34102872e+02 9.41688110e+02 8.69994800e+02 9.65065600e+01\n",
      "  2.53522420e+02 1.23819550e+03 1.38488985e+04 5.50403520e+03\n",
      "  1.00000000e+00]\n",
      " [1.73650700e+01 2.93283800e+01 8.23489700e+00 9.31488400e+00\n",
      "  1.36363600e+01 1.60832400e+01 5.95039700e+01 5.26759800e+00\n",
      "  2.04427800e+01 6.46226400e+01 0.00000000e+00 0.00000000e+00\n",
      "  2.19316851e+02 8.11417560e+02 9.57016600e+02 1.07684350e+02\n",
      "  3.40406090e+02 1.32106640e+03 1.30502233e+04 6.95884680e+03\n",
      "  1.00000000e+00]]\n"
     ]
    }
   ],
   "source": [
    "print(data.exog[:2,:])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Fit and summary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-02-02T06:54:02.714720Z",
     "iopub.status.busy": "2021-02-02T06:54:02.713600Z",
     "iopub.status.idle": "2021-02-02T06:54:02.759498Z",
     "shell.execute_reply": "2021-02-02T06:54:02.760452Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                 Generalized Linear Model Regression Results                  \n",
      "==============================================================================\n",
      "Dep. Variable:           ['y1', 'y2']   No. Observations:                  303\n",
      "Model:                            GLM   Df Residuals:                      282\n",
      "Model Family:                Binomial   Df Model:                           20\n",
      "Link Function:                  logit   Scale:                          1.0000\n",
      "Method:                          IRLS   Log-Likelihood:                -2998.6\n",
      "Date:                Tue, 02 Feb 2021   Deviance:                       4078.8\n",
      "Time:                        06:54:02   Pearson chi2:                 4.05e+03\n",
      "No. Iterations:                     5                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "x1            -0.0168      0.000    -38.749      0.000      -0.018      -0.016\n",
      "x2             0.0099      0.001     16.505      0.000       0.009       0.011\n",
      "x3            -0.0187      0.001    -25.182      0.000      -0.020      -0.017\n",
      "x4            -0.0142      0.000    -32.818      0.000      -0.015      -0.013\n",
      "x5             0.2545      0.030      8.498      0.000       0.196       0.313\n",
      "x6             0.2407      0.057      4.212      0.000       0.129       0.353\n",
      "x7             0.0804      0.014      5.775      0.000       0.053       0.108\n",
      "x8            -1.9522      0.317     -6.162      0.000      -2.573      -1.331\n",
      "x9            -0.3341      0.061     -5.453      0.000      -0.454      -0.214\n",
      "x10           -0.1690      0.033     -5.169      0.000      -0.233      -0.105\n",
      "x11            0.0049      0.001      3.921      0.000       0.002       0.007\n",
      "x12           -0.0036      0.000    -15.878      0.000      -0.004      -0.003\n",
      "x13           -0.0141      0.002     -7.391      0.000      -0.018      -0.010\n",
      "x14           -0.0040      0.000     -8.450      0.000      -0.005      -0.003\n",
      "x15           -0.0039      0.001     -4.059      0.000      -0.006      -0.002\n",
      "x16            0.0917      0.015      6.321      0.000       0.063       0.120\n",
      "x17            0.0490      0.007      6.574      0.000       0.034       0.064\n",
      "x18            0.0080      0.001      5.362      0.000       0.005       0.011\n",
      "x19            0.0002   2.99e-05      7.428      0.000       0.000       0.000\n",
      "x20           -0.0022      0.000     -6.445      0.000      -0.003      -0.002\n",
      "const          2.9589      1.547      1.913      0.056      -0.073       5.990\n",
      "==============================================================================\n"
     ]
    }
   ],
   "source": [
    "glm_binom = sm.GLM(data.endog, data.exog, family=sm.families.Binomial())\n",
    "res = glm_binom.fit()\n",
    "print(res.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Quantities of interest"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-02-02T06:54:02.766478Z",
     "iopub.status.busy": "2021-02-02T06:54:02.765350Z",
     "iopub.status.idle": "2021-02-02T06:54:02.773567Z",
     "shell.execute_reply": "2021-02-02T06:54:02.774547Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total number of trials: 807.0\n",
      "Parameters:  [-1.68150366e-02  9.92547661e-03 -1.87242148e-02 -1.42385609e-02\n",
      "  2.54487173e-01  2.40693664e-01  8.04086739e-02 -1.95216050e+00\n",
      " -3.34086475e-01 -1.69022168e-01  4.91670212e-03 -3.57996435e-03\n",
      " -1.40765648e-02 -4.00499176e-03 -3.90639579e-03  9.17143006e-02\n",
      "  4.89898381e-02  8.04073890e-03  2.22009503e-04 -2.24924861e-03\n",
      "  2.95887793e+00]\n",
      "T-values:  [-38.74908321  16.50473627 -25.1821894  -32.81791308   8.49827113\n",
      "   4.21247925   5.7749976   -6.16191078  -5.45321673  -5.16865445\n",
      "   3.92119964 -15.87825999  -7.39093058  -8.44963886  -4.05916246\n",
      "   6.3210987    6.57434662   5.36229044   7.42806363  -6.44513698\n",
      "   1.91301155]\n"
     ]
    }
   ],
   "source": [
    "print('Total number of trials:',  data.endog[0].sum())\n",
    "print('Parameters: ', res.params)\n",
    "print('T-values: ', res.tvalues)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First differences: We hold all explanatory variables constant at their means and manipulate the percentage of low income households to assess its impact on the response variables: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-02-02T06:54:02.781778Z",
     "iopub.status.busy": "2021-02-02T06:54:02.779764Z",
     "iopub.status.idle": "2021-02-02T06:54:02.788635Z",
     "shell.execute_reply": "2021-02-02T06:54:02.789288Z"
    }
   },
   "outputs": [],
   "source": [
    "means = data.exog.mean(axis=0)\n",
    "means25 = means.copy()\n",
    "means25[0] = stats.scoreatpercentile(data.exog[:,0], 25)\n",
    "means75 = means.copy()\n",
    "means75[0] = lowinc_75per = stats.scoreatpercentile(data.exog[:,0], 75)\n",
    "resp_25 = res.predict(means25)\n",
    "resp_75 = res.predict(means75)\n",
    "diff = resp_75 - resp_25"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The interquartile first difference for the percentage of low income households in a school district is:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-02-02T06:54:02.794572Z",
     "iopub.status.busy": "2021-02-02T06:54:02.793760Z",
     "iopub.status.idle": "2021-02-02T06:54:02.799762Z",
     "shell.execute_reply": "2021-02-02T06:54:02.800419Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "-11.8753%\n"
     ]
    }
   ],
   "source": [
    "print(\"%2.4f%%\" % (diff*100))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plots\n",
    "\n",
    " We extract information that will be used to draw some interesting plots: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-02-02T06:54:02.806704Z",
     "iopub.status.busy": "2021-02-02T06:54:02.805876Z",
     "iopub.status.idle": "2021-02-02T06:54:02.809342Z",
     "shell.execute_reply": "2021-02-02T06:54:02.809981Z"
    }
   },
   "outputs": [],
   "source": [
    "nobs = res.nobs\n",
    "y = data.endog[:,0]/data.endog.sum(1)\n",
    "yhat = res.mu"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plot yhat vs y:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-02-02T06:54:02.815051Z",
     "iopub.status.busy": "2021-02-02T06:54:02.814303Z",
     "iopub.status.idle": "2021-02-02T06:54:02.817322Z",
     "shell.execute_reply": "2021-02-02T06:54:02.818917Z"
    }
   },
   "outputs": [],
   "source": [
    "from statsmodels.graphics.api import abline_plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-02-02T06:54:02.848461Z",
     "iopub.status.busy": "2021-02-02T06:54:02.847287Z",
     "iopub.status.idle": "2021-02-02T06:54:03.205181Z",
     "shell.execute_reply": "2021-02-02T06:54:03.205520Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Fitted values')"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "ax.scatter(yhat, y)\n",
    "line_fit = sm.OLS(y, sm.add_constant(yhat, prepend=True)).fit()\n",
    "abline_plot(model_results=line_fit, ax=ax)\n",
    "\n",
    "\n",
    "ax.set_title('Model Fit Plot')\n",
    "ax.set_ylabel('Observed values')\n",
    "ax.set_xlabel('Fitted values');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Plot yhat vs. Pearson residuals:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-02-02T06:54:03.208974Z",
     "iopub.status.busy": "2021-02-02T06:54:03.207792Z",
     "iopub.status.idle": "2021-02-02T06:54:03.454091Z",
     "shell.execute_reply": "2021-02-02T06:54:03.455069Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Fitted values')"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "\n",
    "ax.scatter(yhat, res.resid_pearson)\n",
    "ax.hlines(0, 0, 1)\n",
    "ax.set_xlim(0, 1)\n",
    "ax.set_title('Residual Dependence Plot')\n",
    "ax.set_ylabel('Pearson Residuals')\n",
    "ax.set_xlabel('Fitted values')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Histogram of standardized deviance residuals:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-02-02T06:54:03.459595Z",
     "iopub.status.busy": "2021-02-02T06:54:03.458203Z",
     "iopub.status.idle": "2021-02-02T06:54:03.775280Z",
     "shell.execute_reply": "2021-02-02T06:54:03.776609Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Histogram of standardized deviance residuals')"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from scipy import stats\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "\n",
    "resid = res.resid_deviance.copy()\n",
    "resid_std = stats.zscore(resid)\n",
    "ax.hist(resid_std, bins=25)\n",
    "ax.set_title('Histogram of standardized deviance residuals');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "QQ Plot of Deviance Residuals:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-02-02T06:54:03.782333Z",
     "iopub.status.busy": "2021-02-02T06:54:03.780696Z",
     "iopub.status.idle": "2021-02-02T06:54:04.233984Z",
     "shell.execute_reply": "2021-02-02T06:54:04.235307Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x576 with 1 Axes>"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from statsmodels import graphics\n",
    "graphics.gofplots.qqplot(resid, line='r')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## GLM: Gamma for proportional count response\n",
    "\n",
    "### Load Scottish Parliament Voting data\n",
    "\n",
    " In the example above, we printed the ``NOTE`` attribute to learn about the\n",
    " Star98 dataset. statsmodels datasets ships with other useful information. For\n",
    " example: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-02-02T06:54:04.241322Z",
     "iopub.status.busy": "2021-02-02T06:54:04.239504Z",
     "iopub.status.idle": "2021-02-02T06:54:04.249197Z",
     "shell.execute_reply": "2021-02-02T06:54:04.250403Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "This data is based on the example in Gill and describes the proportion of\n",
      "voters who voted Yes to grant the Scottish Parliament taxation powers.\n",
      "The data are divided into 32 council districts.  This example's explanatory\n",
      "variables include the amount of council tax collected in pounds sterling as\n",
      "of April 1997 per two adults before adjustments, the female percentage of\n",
      "total claims for unemployment benefits as of January, 1998, the standardized\n",
      "mortality rate (UK is 100), the percentage of labor force participation,\n",
      "regional GDP, the percentage of children aged 5 to 15, and an interaction term\n",
      "between female unemployment and the council tax.\n",
      "\n",
      "The original source files and variable information are included in\n",
      "/scotland/src/\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(sm.datasets.scotland.DESCRLONG)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " Load the data and add a constant to the exogenous variables:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-02-02T06:54:04.256323Z",
     "iopub.status.busy": "2021-02-02T06:54:04.254559Z",
     "iopub.status.idle": "2021-02-02T06:54:04.275877Z",
     "shell.execute_reply": "2021-02-02T06:54:04.277305Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[7.12000e+02 2.10000e+01 1.05000e+02 8.24000e+01 1.35660e+04 1.23000e+01\n",
      "  1.49520e+04 1.00000e+00]\n",
      " [6.43000e+02 2.65000e+01 9.70000e+01 8.02000e+01 1.35660e+04 1.53000e+01\n",
      "  1.70395e+04 1.00000e+00]\n",
      " [6.79000e+02 2.83000e+01 1.13000e+02 8.63000e+01 9.61100e+03 1.39000e+01\n",
      "  1.92157e+04 1.00000e+00]\n",
      " [8.01000e+02 2.71000e+01 1.09000e+02 8.04000e+01 9.48300e+03 1.36000e+01\n",
      "  2.17071e+04 1.00000e+00]\n",
      " [7.53000e+02 2.20000e+01 1.15000e+02 6.47000e+01 9.26500e+03 1.46000e+01\n",
      "  1.65660e+04 1.00000e+00]]\n",
      "[60.3 52.3 53.4 57.  68.7]\n"
     ]
    }
   ],
   "source": [
    "data2 = sm.datasets.scotland.load()\n",
    "data2.exog = sm.add_constant(data2.exog, prepend=False)\n",
    "print(data2.exog[:5,:])\n",
    "print(data2.endog[:5])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model Fit and summary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-02-02T06:54:04.283072Z",
     "iopub.status.busy": "2021-02-02T06:54:04.281307Z",
     "iopub.status.idle": "2021-02-02T06:54:04.309743Z",
     "shell.execute_reply": "2021-02-02T06:54:04.311027Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                 Generalized Linear Model Regression Results                  \n",
      "==============================================================================\n",
      "Dep. Variable:                      y   No. Observations:                   32\n",
      "Model:                            GLM   Df Residuals:                       24\n",
      "Model Family:                   Gamma   Df Model:                            7\n",
      "Link Function:                    log   Scale:                       0.0035927\n",
      "Method:                          IRLS   Log-Likelihood:                -83.110\n",
      "Date:                Tue, 02 Feb 2021   Deviance:                     0.087988\n",
      "Time:                        06:54:04   Pearson chi2:                   0.0862\n",
      "No. Iterations:                     7                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "x1            -0.0024      0.001     -2.466      0.014      -0.004      -0.000\n",
      "x2            -0.1005      0.031     -3.269      0.001      -0.161      -0.040\n",
      "x3             0.0048      0.002      2.946      0.003       0.002       0.008\n",
      "x4            -0.0067      0.003     -2.534      0.011      -0.012      -0.002\n",
      "x5          8.173e-06   7.19e-06      1.136      0.256   -5.93e-06    2.23e-05\n",
      "x6             0.0298      0.015      2.009      0.045       0.001       0.059\n",
      "x7             0.0001   4.33e-05      2.724      0.006    3.31e-05       0.000\n",
      "const          5.6581      0.680      8.318      0.000       4.325       6.991\n",
      "==============================================================================\n"
     ]
    }
   ],
   "source": [
    "glm_gamma = sm.GLM(data2.endog, data2.exog, family=sm.families.Gamma(sm.families.links.log()))\n",
    "glm_results = glm_gamma.fit()\n",
    "print(glm_results.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## GLM: Gaussian distribution with a noncanonical link\n",
    "\n",
    "### Artificial data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-02-02T06:54:04.317508Z",
     "iopub.status.busy": "2021-02-02T06:54:04.315598Z",
     "iopub.status.idle": "2021-02-02T06:54:04.325404Z",
     "shell.execute_reply": "2021-02-02T06:54:04.326726Z"
    }
   },
   "outputs": [],
   "source": [
    "nobs2 = 100\n",
    "x = np.arange(nobs2)\n",
    "np.random.seed(54321)\n",
    "X = np.column_stack((x,x**2))\n",
    "X = sm.add_constant(X, prepend=False)\n",
    "lny = np.exp(-(.03*x + .0001*x**2 - 1.0)) + .001 * np.random.rand(nobs2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Fit and summary (artificial data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "execution": {
     "iopub.execute_input": "2021-02-02T06:54:04.332497Z",
     "iopub.status.busy": "2021-02-02T06:54:04.330705Z",
     "iopub.status.idle": "2021-02-02T06:54:04.355111Z",
     "shell.execute_reply": "2021-02-02T06:54:04.356426Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                 Generalized Linear Model Regression Results                  \n",
      "==============================================================================\n",
      "Dep. Variable:                      y   No. Observations:                  100\n",
      "Model:                            GLM   Df Residuals:                       97\n",
      "Model Family:                Gaussian   Df Model:                            2\n",
      "Link Function:                    log   Scale:                      1.0531e-07\n",
      "Method:                          IRLS   Log-Likelihood:                 662.92\n",
      "Date:                Tue, 02 Feb 2021   Deviance:                   1.0215e-05\n",
      "Time:                        06:54:04   Pearson chi2:                 1.02e-05\n",
      "No. Iterations:                     7                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "x1            -0.0300    5.6e-06  -5361.316      0.000      -0.030      -0.030\n",
      "x2         -9.939e-05   1.05e-07   -951.091      0.000   -9.96e-05   -9.92e-05\n",
      "const          1.0003   5.39e-05   1.86e+04      0.000       1.000       1.000\n",
      "==============================================================================\n"
     ]
    }
   ],
   "source": [
    "gauss_log = sm.GLM(lny, X, family=sm.families.Gaussian(sm.families.links.log()))\n",
    "gauss_log_results = gauss_log.fit()\n",
    "print(gauss_log_results.summary())"
   ]
  }
 ],
 "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": "3.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}