{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "bda38497",
   "metadata": {},
   "source": [
    "## Hurdle and truncated count models\n",
    "\n",
    "Author: Josef Perktold\n",
    "\n",
    "Statsmodels has now hurdle and truncated count models, added in version 0.14.\n",
    "\n",
    "A hurdle model is composed of a model for zeros and a model for the distribution for counts larger than zero. The zero model is a binary model for a count of zero versus larger than zero. The count model for nonzero counts is a zero truncated count model.\n",
    "\n",
    "Statsmodels currently supports hurdle models with Poisson and Negative Binomial distributions as zero model and as count model. Binary models like Logit, Probit or GLM-Binomial are not yet supported as zero model.\n",
    "The advantage of Poisson-Poisson hurdle is that the standard Poisson model is a special case with equal parameters in both models. This provides a simple Wald test for the hurdle model against the Poisson model.\n",
    "\n",
    "The implemented binary model is a censored model where observations are right censored at one. That means that only 0 or 1 counts are observed.\n",
    "\n",
    "The hurdle model can be estimated by separately estimating the zero model and the count model for the zero truncated data assuming that observations are independently distributed (no correlation across observations). The resulting covariance matrix of the parameter estimates is block diagonal with diagonal blocks given by the submodels.\n",
    "Joint estimation is not yet implemented.\n",
    "\n",
    "The censored and truncated count models were developed mainly to support the hurdle model. However, the left truncated count models have other applications than supporting the hurdle models. The right censored models are not of separate interest because they only support binary observations that can be modeled by GLM-Binomial, Logit or Probit.\n",
    "\n",
    "For the hurdle model there is a single class `HurdleCountModel`, that includes the distributions of the submodels as option. \n",
    "Classes for truncated models are currently `TruncatedLFPoisson` and `TruncatedLFNegativeBinomialP`, where \"LF\" stands for left truncation at a fixed, observation independent truncation point. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "29e6105a",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "eed890e6",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-19T16:41:33.372545Z",
     "iopub.status.busy": "2024-04-19T16:41:33.372163Z",
     "iopub.status.idle": "2024-04-19T16:41:36.036844Z",
     "shell.execute_reply": "2024-04-19T16:41:36.036142Z"
    }
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "import statsmodels.discrete.truncated_model as smtc\n",
    "\n",
    "from statsmodels.discrete.discrete_model import (\n",
    "    Poisson, NegativeBinomial, NegativeBinomialP, GeneralizedPoisson)\n",
    "from statsmodels.discrete.count_model import (\n",
    "    ZeroInflatedPoisson,\n",
    "    ZeroInflatedGeneralizedPoisson,\n",
    "    ZeroInflatedNegativeBinomialP\n",
    "    )\n",
    "\n",
    "from statsmodels.discrete.truncated_model import (\n",
    "    TruncatedLFPoisson,\n",
    "    TruncatedLFNegativeBinomialP,\n",
    "    _RCensoredPoisson,\n",
    "    HurdleCountModel,\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "85da298e",
   "metadata": {},
   "source": [
    "## Simulating a hurdle model\n",
    "\n",
    "We are simulating a Poisson-Poisson hurdle model explicitly because there are not yet any distribution helper functions for it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "99c1d162",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-19T16:41:36.043165Z",
     "iopub.status.busy": "2024-04-19T16:41:36.042725Z",
     "iopub.status.idle": "2024-04-19T16:41:36.090356Z",
     "shell.execute_reply": "2024-04-19T16:41:36.086710Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[102 335 590 770 816 739 573 402 265 176 116  59  35   7  11   4]\n",
      "4898\n",
      "602\n",
      "93\n",
      "11\n",
      "2\n",
      "0\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([ 102, 1448, 1502, 1049,  542,  234,   81,   31,    6,    5])"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.random.seed(987456348)\n",
    "# large sample to get strong results\n",
    "nobs = 5000\n",
    "x = np.column_stack((np.ones(nobs), np.linspace(0, 1, nobs)))\n",
    "\n",
    "mu0 = np.exp(0.5 *2 * x.sum(1))\n",
    "y = np.random.poisson(mu0, size=nobs)\n",
    "print(np.bincount(y))\n",
    "y_ = y\n",
    "indices = np.arange(len(y))\n",
    "mask = mask0 = y > 0 \n",
    "for _ in range(10):\n",
    "    \n",
    "    print( mask.sum())\n",
    "    indices = mask #indices[mask]\n",
    "    if not np.any(mask):\n",
    "        break\n",
    "    mu_ = np.exp(0.5 * x[indices].sum(1))\n",
    "    y[indices] = y_ = np.random.poisson(mu_, size=len(mu_))\n",
    "    np.place(y, mask, y_)\n",
    "    mask = np.logical_and(mask0, y == 0)\n",
    "    \n",
    "np.bincount(y)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b6435d35",
   "metadata": {},
   "source": [
    "## Estimating misspecified Poisson Model\n",
    "\n",
    "The data that we generated has zero deflation, this is, we observe fewer zeros than what we would expect in a Poisson model.\n",
    "\n",
    "After fitting the model, we can use the plot function in the poisson diagnostic class to compare the expected predictive distribution and the realized frequencies. The shows that the Poisson model overestimates the number of zeros and underestimates counts of one and two."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "7fdd59b7",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-19T16:41:36.093559Z",
     "iopub.status.busy": "2024-04-19T16:41:36.093289Z",
     "iopub.status.idle": "2024-04-19T16:41:36.206152Z",
     "shell.execute_reply": "2024-04-19T16:41:36.203157Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Optimization terminated successfully.\n",
      "         Current function value: 1.668079\n",
      "         Iterations 4\n",
      "                          Poisson Regression Results                          \n",
      "==============================================================================\n",
      "Dep. Variable:                      y   No. Observations:                 5000\n",
      "Model:                        Poisson   Df Residuals:                     4998\n",
      "Method:                           MLE   Df Model:                            1\n",
      "Date:                Fri, 19 Apr 2024   Pseudo R-squ.:                0.008678\n",
      "Time:                        16:41:36   Log-Likelihood:                -8340.4\n",
      "converged:                       True   LL-Null:                       -8413.4\n",
      "Covariance Type:            nonrobust   LLR p-value:                 1.279e-33\n",
      "==============================================================================\n",
      "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "const          0.6532      0.019     33.642      0.000       0.615       0.691\n",
      "x1             0.3871      0.032     12.062      0.000       0.324       0.450\n",
      "==============================================================================\n"
     ]
    }
   ],
   "source": [
    "mod_p = Poisson(y, x)\n",
    "res_p = mod_p.fit()\n",
    "print(res_p.summary())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "91a5c9cb",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-19T16:41:36.209390Z",
     "iopub.status.busy": "2024-04-19T16:41:36.209114Z",
     "iopub.status.idle": "2024-04-19T16:41:39.417525Z",
     "shell.execute_reply": "2024-04-19T16:41:39.416722Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x1200 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "dia_p = res_p.get_diagnostic()\n",
    "dia_p.plot_probs();"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a237e21f",
   "metadata": {},
   "source": [
    "## Estimating the Hurdle Model\n",
    "\n",
    "Next, we estimate the correctly specified Poisson-Poisson hurdle model.\n",
    "\n",
    "Signature and options for the HurdleCountModel shows that poisson-poisson is the default, so we do not need to specify any options when creating this model.\n",
    "\n",
    "`HurdleCountModel(endog, exog, offset=None, dist='poisson', zerodist='poisson', \n",
    "                  p=2, pzero=2, exposure=None, missing='none', **kwargs)`\n",
    "                  \n",
    "The results class of the HurdleCountModel has a `get_diagnostic` method. However, only part of the diagnostic methods are currently available. The plot of the predictive distribution shows very high agreement with the data.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "70065ce8",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-19T16:41:39.420551Z",
     "iopub.status.busy": "2024-04-19T16:41:39.420162Z",
     "iopub.status.idle": "2024-04-19T16:41:39.997344Z",
     "shell.execute_reply": "2024-04-19T16:41:39.996659Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                     HurdleCountModel Regression Results                      \n",
      "==============================================================================\n",
      "Dep. Variable:                      y   No. Observations:                 5000\n",
      "Model:               HurdleCountModel   Df Residuals:                     4996\n",
      "Method:                           MLE   Df Model:                            2\n",
      "Date:                Fri, 19 Apr 2024   Pseudo R-squ.:                 0.01503\n",
      "Time:                        16:41:39   Log-Likelihood:                -8004.9\n",
      "converged:               [True, True]   LL-Null:                       -8127.1\n",
      "Covariance Type:            nonrobust   LLR p-value:                 8.901e-54\n",
      "==============================================================================\n",
      "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "zm_const       0.9577      0.048     20.063      0.000       0.864       1.051\n",
      "zm_x1          1.0576      0.121      8.737      0.000       0.820       1.295\n",
      "const          0.5009      0.024     20.875      0.000       0.454       0.548\n",
      "x1             0.4577      0.039     11.882      0.000       0.382       0.533\n",
      "==============================================================================\n"
     ]
    }
   ],
   "source": [
    "mod_h = HurdleCountModel(y, x)\n",
    "res_h = mod_h.fit(disp=False)\n",
    "print(res_h.summary())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "6c3aadbf",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-19T16:41:40.002608Z",
     "iopub.status.busy": "2024-04-19T16:41:40.001330Z",
     "iopub.status.idle": "2024-04-19T16:41:42.331600Z",
     "shell.execute_reply": "2024-04-19T16:41:42.330931Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x1200 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "dia_h = res_h.get_diagnostic()\n",
    "dia_h.plot_probs();"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0827fde3",
   "metadata": {},
   "source": [
    "We can use the Wald test to test whether the parameters of the zero model are the same as the parameters of the zero-truncated count model. The p-value is very small and correctly rejects that the model is just Poisson. We are using a large sample size, so the power of the test will be large in this case."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "c000b4cb",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-19T16:41:42.336092Z",
     "iopub.status.busy": "2024-04-19T16:41:42.334990Z",
     "iopub.status.idle": "2024-04-19T16:41:42.359678Z",
     "shell.execute_reply": "2024-04-19T16:41:42.359036Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<class 'statsmodels.stats.contrast.ContrastResults'>\n",
       "<Wald test (chi2): statistic=470.67320754391915, p-value=6.231772522807044e-103, df_denom=2>"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "res_h.wald_test(\"zm_const = const, zm_x1 = x1\", scalar=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "100ba8a8",
   "metadata": {},
   "source": [
    "## Prediction\n",
    "\n",
    "The hurdle model can be used for prediction for statistics of the overall model and of the two submodels. The statistics that should be predicted is specified using the `which` keyword.\n",
    "\n",
    "The following is taken from the docstring for predict and lists available the options.\n",
    "\n",
    "        which : str (optional)\n",
    "            Statitistic to predict. Default is 'mean'.\n",
    "\n",
    "            - 'mean' : the conditional expectation of endog E(y | x)\n",
    "            - 'mean-main' : mean parameter of truncated count model.\n",
    "              Note, this is not the mean of the truncated distribution.\n",
    "            - 'linear' : the linear predictor of the truncated count model.\n",
    "            - 'var' : returns the estimated variance of endog implied by the\n",
    "              model.\n",
    "            - 'prob-main' : probability of selecting the main model which is\n",
    "              the probability of observing a nonzero count P(y > 0 | x).\n",
    "            - 'prob-zero' : probability of observing a zero count. P(y=0 | x).\n",
    "              This is equal to is ``1 - prob-main``\n",
    "            - 'prob-trunc' : probability of truncation of the truncated count\n",
    "              model. This is the probability of observing a zero count implied\n",
    "              by the truncation model.\n",
    "            - 'mean-nonzero' : expected value conditional on having observation\n",
    "              larger than zero, E(y | X, y>0)\n",
    "            - 'prob' : probabilities of each count from 0 to max(endog), or\n",
    "              for y_values if those are provided. This is a multivariate\n",
    "              return (2-dim when predicting for several observations).\n",
    "              \n",
    "These options are available in the `predict` and the `get_prediction` methods of the results class.\n",
    "\n",
    "For the following example, we create a set of explanatory variables that are taken from the original data at equal spaced intervals. Then we can predict the available statistics conditional on these explanatory variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "9757da68",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-19T16:41:42.364110Z",
     "iopub.status.busy": "2024-04-19T16:41:42.363049Z",
     "iopub.status.idle": "2024-04-19T16:41:42.382766Z",
     "shell.execute_reply": "2024-04-19T16:41:42.382139Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1.        , 0.        ],\n",
       "       [1.        , 0.20004001],\n",
       "       [1.        , 0.40008002],\n",
       "       [1.        , 0.60012002],\n",
       "       [1.        , 0.80016003]])"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "which_options = [\"mean\", \"mean-main\", \"linear\", \"mean-nonzero\", \"prob-zero\", \"prob-main\", \"prob-trunc\", \"var\", \"prob\"]\n",
    "ex = x[slice(None, None, nobs // 5), :]\n",
    "ex"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "3614ee25",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-19T16:41:42.387378Z",
     "iopub.status.busy": "2024-04-19T16:41:42.386218Z",
     "iopub.status.idle": "2024-04-19T16:41:42.410786Z",
     "shell.execute_reply": "2024-04-19T16:41:42.410140Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean\n",
      "     [1.89150663 2.07648059 2.25555158 2.43319456 2.61673457]\n",
      "mean-main\n",
      "     [1.65015181 1.8083782  1.98177629 2.17180081 2.38004602]\n",
      "linear\n",
      "     [0.50086729 0.59243042 0.68399356 0.77555669 0.86711982]\n",
      "mean-nonzero\n",
      "     [2.04231955 2.16292424 2.29857565 2.45116551 2.62277411]\n",
      "prob-zero\n",
      "     [0.07384394 0.0399661  0.01871771 0.00733159 0.00230273]\n",
      "prob-main\n",
      "     [0.92615606 0.9600339  0.98128229 0.99266841 0.99769727]\n",
      "prob-trunc\n",
      "     [0.19202076 0.16391977 0.1378242  0.11397219 0.09254632]\n",
      "var\n",
      "     [1.43498239 1.51977118 1.63803729 1.7971727  1.99738345]\n",
      "prob\n",
      "     [[7.38439416e-02 3.63208532e-01 2.99674608e-01 1.64836199e-01\n",
      "  6.80011882e-02 2.24424568e-02 6.17224344e-03 1.45501981e-03\n",
      "  3.00125448e-04 5.50280612e-05]\n",
      " [3.99660987e-02 3.40376213e-01 3.07764462e-01 1.85518182e-01\n",
      "  8.38717591e-02 3.03343722e-02 9.14266959e-03 2.36191491e-03\n",
      "  5.33904431e-04 1.07277904e-04]\n",
      " [1.87177088e-02 3.10869602e-01 3.08037002e-01 2.03486809e-01\n",
      "  1.00816333e-01 3.99590837e-02 1.31983274e-02 3.73659033e-03\n",
      "  9.25635762e-04 2.03822556e-04]\n",
      " [7.33159258e-03 2.77316512e-01 3.01138113e-01 2.18003999e-01\n",
      "  1.18365316e-01 5.14131777e-02 1.86098635e-02 5.77384524e-03\n",
      "  1.56745522e-03 3.78244503e-04]\n",
      " [2.30272798e-03 2.42169151e-01 2.88186862e-01 2.28632665e-01\n",
      "  1.36039066e-01 6.47558475e-02 2.56869828e-02 8.73374304e-03\n",
      "  2.59833880e-03 6.87129546e-04]]\n"
     ]
    }
   ],
   "source": [
    "for w in which_options:\n",
    "    print(w)\n",
    "    pred = res_h.predict(ex, which=w)\n",
    "    print(\"    \", pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "bbc9ebed",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-19T16:41:42.415250Z",
     "iopub.status.busy": "2024-04-19T16:41:42.414191Z",
     "iopub.status.idle": "2024-04-19T16:41:42.445648Z",
     "shell.execute_reply": "2024-04-19T16:41:42.444971Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean\n",
      "     [1.89150663 2.07648059 2.25555158 2.43319456 2.61673457]\n",
      "  se [0.07877461 0.05693768 0.05866892 0.09551274 0.15359057]\n",
      "mean-main\n",
      "     [1.65015181 1.8083782  1.98177629 2.17180081 2.38004602]\n",
      "  se [0.03959242 0.03164634 0.02471869 0.02415162 0.03453261]\n",
      "linear\n",
      "     [0.50086729 0.59243042 0.68399356 0.77555669 0.86711982]\n",
      "  se [0.04773779 0.03148549 0.02960421 0.04397859 0.06453261]\n",
      "mean-nonzero\n",
      "     [2.04231955 2.16292424 2.29857565 2.45116551 2.62277411]\n",
      "  se [0.02978486 0.02443098 0.01958745 0.0196433  0.02881753]\n",
      "prob-zero\n",
      "     [0.07384394 0.0399661  0.01871771 0.00733159 0.00230273]\n",
      "  se [0.00918583 0.00405155 0.00220446 0.00158494 0.00090255]\n",
      "prob-main\n",
      "     [0.92615606 0.9600339  0.98128229 0.99266841 0.99769727]\n",
      "  se [0.00918583 0.00405155 0.00220446 0.00158494 0.00090255]\n",
      "prob-trunc\n",
      "     [0.19202076 0.16391977 0.1378242  0.11397219 0.09254632]\n",
      "  se [0.00760257 0.00518746 0.00340683 0.00275261 0.00319587]\n",
      "var\n",
      "     [1.43498239 1.51977118 1.63803729 1.7971727  1.99738345]\n",
      "  se [0.04853902 0.03615054 0.02747485 0.02655145 0.03733328]\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/hostedtoolcache/Python/3.10.14/x64/lib/python3.10/site-packages/statsmodels/base/_prediction_inference.py:782: UserWarning: using default log-link in get_prediction\n",
      "  warnings.warn(\"using default log-link in get_prediction\")\n"
     ]
    }
   ],
   "source": [
    "for w in which_options[:-1]:\n",
    "    print(w)\n",
    "    pred = res_h.get_prediction(ex, which=w)\n",
    "    print(\"    \", pred.predicted)\n",
    "    print(\"  se\", pred.se)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e356eb1b",
   "metadata": {},
   "source": [
    "The option `which=\"prob\"` returns an array of predicted probabilities for each row of the predict `exog`.\n",
    "We are often interested in the mean probabilities averaged over all exog. The prediction methods have an option `average=True` to compute the average of the predicted values across observations and the corresponding standard errors and confidence intervals for those averaged predictions."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "5d90afc2",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-19T16:41:42.450177Z",
     "iopub.status.busy": "2024-04-19T16:41:42.449099Z",
     "iopub.status.idle": "2024-04-19T16:41:42.470226Z",
     "shell.execute_reply": "2024-04-19T16:41:42.469578Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     [2.84324139e-02 3.06788002e-01 3.00960210e-01 2.00095571e-01\n",
      " 1.01418732e-01 4.17809876e-02 1.45620174e-02 4.41222267e-03\n",
      " 1.18509193e-03 2.86300514e-04]\n",
      "  se [2.81472152e-03 5.00830805e-03 1.37524763e-03 1.87343644e-03\n",
      " 1.99068649e-03 1.23878525e-03 5.78099173e-04 2.21180110e-04\n",
      " 7.25021189e-05 2.08872558e-05]\n"
     ]
    }
   ],
   "source": [
    "pred = res_h.get_prediction(ex, which=\"prob\", average=True)\n",
    "print(\"    \", pred.predicted)\n",
    "print(\"  se\", pred.se)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "09602e97",
   "metadata": {},
   "source": [
    "We use the panda DataFrame to get a display that is easier to read. The \"predicted\" column shows the probability mass function for the predicted distribution of response values averaged of our 5 grid points of exog. The probabilities do not add up to one because counts larger than those observed have positive probability and are missing in the table, although in this example that probability is small."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "77c7eff9",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-19T16:41:42.473622Z",
     "iopub.status.busy": "2024-04-19T16:41:42.473364Z",
     "iopub.status.idle": "2024-04-19T16:41:42.502167Z",
     "shell.execute_reply": "2024-04-19T16:41:42.501500Z"
    }
   },
   "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>predicted</th>\n",
       "      <th>se</th>\n",
       "      <th>ci_lower</th>\n",
       "      <th>ci_upper</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.028432</td>\n",
       "      <td>0.002815</td>\n",
       "      <td>0.022916</td>\n",
       "      <td>0.033949</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.306788</td>\n",
       "      <td>0.005008</td>\n",
       "      <td>0.296972</td>\n",
       "      <td>0.316604</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.300960</td>\n",
       "      <td>0.001375</td>\n",
       "      <td>0.298265</td>\n",
       "      <td>0.303656</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.200096</td>\n",
       "      <td>0.001873</td>\n",
       "      <td>0.196424</td>\n",
       "      <td>0.203767</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.101419</td>\n",
       "      <td>0.001991</td>\n",
       "      <td>0.097517</td>\n",
       "      <td>0.105320</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.041781</td>\n",
       "      <td>0.001239</td>\n",
       "      <td>0.039353</td>\n",
       "      <td>0.044209</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.014562</td>\n",
       "      <td>0.000578</td>\n",
       "      <td>0.013429</td>\n",
       "      <td>0.015695</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.004412</td>\n",
       "      <td>0.000221</td>\n",
       "      <td>0.003979</td>\n",
       "      <td>0.004846</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.001185</td>\n",
       "      <td>0.000073</td>\n",
       "      <td>0.001043</td>\n",
       "      <td>0.001327</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.000286</td>\n",
       "      <td>0.000021</td>\n",
       "      <td>0.000245</td>\n",
       "      <td>0.000327</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   predicted        se  ci_lower  ci_upper\n",
       "0   0.028432  0.002815  0.022916  0.033949\n",
       "1   0.306788  0.005008  0.296972  0.316604\n",
       "2   0.300960  0.001375  0.298265  0.303656\n",
       "3   0.200096  0.001873  0.196424  0.203767\n",
       "4   0.101419  0.001991  0.097517  0.105320\n",
       "5   0.041781  0.001239  0.039353  0.044209\n",
       "6   0.014562  0.000578  0.013429  0.015695\n",
       "7   0.004412  0.000221  0.003979  0.004846\n",
       "8   0.001185  0.000073  0.001043  0.001327\n",
       "9   0.000286  0.000021  0.000245  0.000327"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfp_h = pred.summary_frame()\n",
    "dfp_h"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "185a5042",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-19T16:41:42.506612Z",
     "iopub.status.busy": "2024-04-19T16:41:42.505477Z",
     "iopub.status.idle": "2024-04-19T16:41:42.519969Z",
     "shell.execute_reply": "2024-04-19T16:41:42.519328Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0.9999215487936677, 7.84512063323195e-05)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prob_larger9 = pred.predicted.sum()\n",
    "prob_larger9, 1 - prob_larger9"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a75795cd",
   "metadata": {},
   "source": [
    "`get_prediction` returns in this case an instance of the base `PredictionResultsDelta` class.\n",
    "\n",
    "Inferential statistics like standard errors, p-values and confidence interval for nonlinear functions that depend on several distribution parameters are computed using the delta method. Inference for predictions is based on the normal distribution. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "f8124e02",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-19T16:41:42.524460Z",
     "iopub.status.busy": "2024-04-19T16:41:42.523363Z",
     "iopub.status.idle": "2024-04-19T16:41:42.538820Z",
     "shell.execute_reply": "2024-04-19T16:41:42.538143Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<statsmodels.base._prediction_inference.PredictionResultsDelta at 0x7f546b687760>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "6c1d314d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-19T16:41:42.543101Z",
     "iopub.status.busy": "2024-04-19T16:41:42.542016Z",
     "iopub.status.idle": "2024-04-19T16:41:42.555639Z",
     "shell.execute_reply": "2024-04-19T16:41:42.555016Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(<scipy.stats._continuous_distns.norm_gen at 0x7f5471f837c0>, ())"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred.dist, pred.dist_args"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6b032002",
   "metadata": {},
   "source": [
    "We can compare the distribution predicted by the hurdle model with the one predicted by the Poisson model that we estimated earlier. The last column, \"diff\", shows that Poisson model overestimates the number of zeros by around 8% of observations and underestimates the counts of 1 and 2 by 7%, resp. 3.7% at the average over the `exog` grid."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "dd01ba63",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-19T16:41:42.559961Z",
     "iopub.status.busy": "2024-04-19T16:41:42.558896Z",
     "iopub.status.idle": "2024-04-19T16:41:42.598920Z",
     "shell.execute_reply": "2024-04-19T16:41:42.598188Z"
    }
   },
   "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>predicted</th>\n",
       "      <th>se</th>\n",
       "      <th>ci_lower</th>\n",
       "      <th>ci_upper</th>\n",
       "      <th>poisson</th>\n",
       "      <th>diff</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.028432</td>\n",
       "      <td>0.002815</td>\n",
       "      <td>0.022916</td>\n",
       "      <td>0.033949</td>\n",
       "      <td>0.107848</td>\n",
       "      <td>0.079416</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.306788</td>\n",
       "      <td>0.005008</td>\n",
       "      <td>0.296972</td>\n",
       "      <td>0.316604</td>\n",
       "      <td>0.237020</td>\n",
       "      <td>-0.069768</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.300960</td>\n",
       "      <td>0.001375</td>\n",
       "      <td>0.298265</td>\n",
       "      <td>0.303656</td>\n",
       "      <td>0.263523</td>\n",
       "      <td>-0.037437</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.200096</td>\n",
       "      <td>0.001873</td>\n",
       "      <td>0.196424</td>\n",
       "      <td>0.203767</td>\n",
       "      <td>0.197657</td>\n",
       "      <td>-0.002439</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.101419</td>\n",
       "      <td>0.001991</td>\n",
       "      <td>0.097517</td>\n",
       "      <td>0.105320</td>\n",
       "      <td>0.112511</td>\n",
       "      <td>0.011093</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.041781</td>\n",
       "      <td>0.001239</td>\n",
       "      <td>0.039353</td>\n",
       "      <td>0.044209</td>\n",
       "      <td>0.051833</td>\n",
       "      <td>0.010052</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.014562</td>\n",
       "      <td>0.000578</td>\n",
       "      <td>0.013429</td>\n",
       "      <td>0.015695</td>\n",
       "      <td>0.020124</td>\n",
       "      <td>0.005561</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.004412</td>\n",
       "      <td>0.000221</td>\n",
       "      <td>0.003979</td>\n",
       "      <td>0.004846</td>\n",
       "      <td>0.006769</td>\n",
       "      <td>0.002356</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.001185</td>\n",
       "      <td>0.000073</td>\n",
       "      <td>0.001043</td>\n",
       "      <td>0.001327</td>\n",
       "      <td>0.002012</td>\n",
       "      <td>0.000827</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.000286</td>\n",
       "      <td>0.000021</td>\n",
       "      <td>0.000245</td>\n",
       "      <td>0.000327</td>\n",
       "      <td>0.000537</td>\n",
       "      <td>0.000250</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   predicted        se  ci_lower  ci_upper   poisson      diff\n",
       "0   0.028432  0.002815  0.022916  0.033949  0.107848  0.079416\n",
       "1   0.306788  0.005008  0.296972  0.316604  0.237020 -0.069768\n",
       "2   0.300960  0.001375  0.298265  0.303656  0.263523 -0.037437\n",
       "3   0.200096  0.001873  0.196424  0.203767  0.197657 -0.002439\n",
       "4   0.101419  0.001991  0.097517  0.105320  0.112511  0.011093\n",
       "5   0.041781  0.001239  0.039353  0.044209  0.051833  0.010052\n",
       "6   0.014562  0.000578  0.013429  0.015695  0.020124  0.005561\n",
       "7   0.004412  0.000221  0.003979  0.004846  0.006769  0.002356\n",
       "8   0.001185  0.000073  0.001043  0.001327  0.002012  0.000827\n",
       "9   0.000286  0.000021  0.000245  0.000327  0.000537  0.000250"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pred_p = res_p.get_prediction(ex, which=\"prob\", average=True)\n",
    "dfp_p = pred_p.summary_frame()\n",
    "dfp_h[\"poisson\"] = dfp_p[\"predicted\"]\n",
    "dfp_h[\"diff\"] = dfp_h[\"poisson\"] - dfp_h[\"predicted\"]\n",
    "dfp_h"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0da7fb38",
   "metadata": {},
   "source": [
    "## Other post-estimation\n",
    "\n",
    "The estimated hurdle model can be use for wald test of parameters and for prediction. Other maximum likelihood statistics such as loglikelihood value and information criteria are also available. \n",
    "\n",
    "However, some post-estimation methods that require helper functions that are not needed for estimation, parameter inference and prediction are not yet available. The main methods that are not supported yet are `score_test`, `get_distribution`, and `get_influence`. Diagnostic measures in `get_diagnostics` are only available for statistics that are based on prediction.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "e821f6e8",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-19T16:41:42.616095Z",
     "iopub.status.busy": "2024-04-19T16:41:42.614615Z",
     "iopub.status.idle": "2024-04-19T16:41:42.622408Z",
     "shell.execute_reply": "2024-04-19T16:41:42.621780Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(-8004.904002793644, 4996, 16017.808005587289, 16043.876778352953)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "res_h.llf, res_h.df_resid, res_h.aic, res_h.bic"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "31d2ae62",
   "metadata": {},
   "source": [
    "Is there excess dispersion? We can use the pearson residuals to compute a pearson chi2 statistics which should be close to 1 if the model is correctly specified."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "9f1a5124",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-19T16:41:42.625788Z",
     "iopub.status.busy": "2024-04-19T16:41:42.625546Z",
     "iopub.status.idle": "2024-04-19T16:41:42.634186Z",
     "shell.execute_reply": "2024-04-19T16:41:42.633520Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9989670114949286"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(res_h.resid_pearson**2).sum() / res_h.df_resid"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9ff9df60",
   "metadata": {},
   "source": [
    "The diagnostic class also has the predictive distribution which is used in the diagnostic plots. No other statistics or tests are currently availalbe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "50ba5545",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-19T16:41:42.638672Z",
     "iopub.status.busy": "2024-04-19T16:41:42.637521Z",
     "iopub.status.idle": "2024-04-19T16:41:42.650808Z",
     "shell.execute_reply": "2024-04-19T16:41:42.650136Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.02044612, 0.29147174, 0.29856288, 0.20740118, 0.10990976,\n",
       "       0.04737579, 0.0172898 , 0.00548983, 0.00154646, 0.00039214])"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dia_h.probs_predicted.mean(0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "c913a1bb",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-04-19T16:41:42.655242Z",
     "iopub.status.busy": "2024-04-19T16:41:42.654083Z",
     "iopub.status.idle": "2024-04-19T16:41:42.671462Z",
     "shell.execute_reply": "2024-04-19T16:41:42.670834Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 1.10849337,  1.10830496, -0.89188344, -0.89207183,  1.10773978,\n",
       "       -0.8924486 , -0.89263697,  0.10717466,  0.1069863 ,  0.10679794])"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "res_h.resid[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7f4f45d2",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.10.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}