5_Modelos_de_Clasificacion.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "5 Modelos de Clasificacion.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyOj7HwUyQNm8Ou1iyXGEpd/",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/rpasquini/herramientas-econometricas/blob/main/5_Modelos_de_Clasificacion.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "35RIk2hM17u2",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "1e7cfd10-2a8b-409e-9b7e-122bb557609b"
      },
      "source": [
        "!pip install geopandas==0.13.2\n",
        "\n",
        "import geopandas as gpd\n",
        "import pandas as pd\n",
        "import statsmodels.api as sm\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: geopandas==0.13.2 in /usr/local/lib/python3.10/dist-packages (0.13.2)\n",
            "Requirement already satisfied: fiona>=1.8.19 in /usr/local/lib/python3.10/dist-packages (from geopandas==0.13.2) (1.9.4.post1)\n",
            "Requirement already satisfied: packaging in /usr/local/lib/python3.10/dist-packages (from geopandas==0.13.2) (23.1)\n",
            "Requirement already satisfied: pandas>=1.1.0 in /usr/local/lib/python3.10/dist-packages (from geopandas==0.13.2) (1.5.3)\n",
            "Requirement already satisfied: pyproj>=3.0.1 in /usr/local/lib/python3.10/dist-packages (from geopandas==0.13.2) (3.6.0)\n",
            "Requirement already satisfied: shapely>=1.7.1 in /usr/local/lib/python3.10/dist-packages (from geopandas==0.13.2) (2.0.1)\n",
            "Requirement already satisfied: attrs>=19.2.0 in /usr/local/lib/python3.10/dist-packages (from fiona>=1.8.19->geopandas==0.13.2) (23.1.0)\n",
            "Requirement already satisfied: certifi in /usr/local/lib/python3.10/dist-packages (from fiona>=1.8.19->geopandas==0.13.2) (2023.5.7)\n",
            "Requirement already satisfied: click~=8.0 in /usr/local/lib/python3.10/dist-packages (from fiona>=1.8.19->geopandas==0.13.2) (8.1.4)\n",
            "Requirement already satisfied: click-plugins>=1.0 in /usr/local/lib/python3.10/dist-packages (from fiona>=1.8.19->geopandas==0.13.2) (1.1.1)\n",
            "Requirement already satisfied: cligj>=0.5 in /usr/local/lib/python3.10/dist-packages (from fiona>=1.8.19->geopandas==0.13.2) (0.7.2)\n",
            "Requirement already satisfied: six in /usr/local/lib/python3.10/dist-packages (from fiona>=1.8.19->geopandas==0.13.2) (1.16.0)\n",
            "Requirement already satisfied: python-dateutil>=2.8.1 in /usr/local/lib/python3.10/dist-packages (from pandas>=1.1.0->geopandas==0.13.2) (2.8.2)\n",
            "Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas>=1.1.0->geopandas==0.13.2) (2022.7.1)\n",
            "Requirement already satisfied: numpy>=1.21.0 in /usr/local/lib/python3.10/dist-packages (from pandas>=1.1.0->geopandas==0.13.2) (1.22.4)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-gmLupim_U2m"
      },
      "source": [
        "## Estimación de un Modelo de Elección de \"Ubicación Residencial\" adaptado a una situación de vivir en la formalidad vs. la informalidad.\n",
        "\n",
        "A modo de ejemplo, vamos a intentar explicar predecir la localización de los hogares en las áreas de la \"ciudad formal\" o en los \"asentamientos informales\". Vamos a utilizar un extracto de los datos de la investigación de Goytia, Heikkila, Pasquini (2020).\n",
        "   "
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Descripción de variables\n",
        "* bformal es una dummy que señala con 1 a las areas \"formales\" y 0 a las areas \"informales\". Este es un criterio de delimitación espacial.  \"Informal\" se utiliza para el área que constituye un barrio popular  \n",
        "* dconurb1, dconurb2, dconurb3 son dummies que señalan a los agrupamientos de partidos del AMBA \"Conurbano 1\", \"Conurbano 2\" y \"Conurbano 3\" respectivamente, según la categorización del INDEC.  Vease https://www.indec.gob.ar/dbindec/folleto_gba.pdf\n",
        "* dprop_soloviv', 'dprop_ocupant', 'dinquilinos' son dummies que señalan a los que son propietarios solo de la vivienda, ocupantes e inquilinos respectivamente.\n",
        "* dedu_secincomp', 'dedu_seccomp', 'dedu_ter_o_mas' son dummies que señalan la educación del jefe de hogar\n",
        "* 'dconexionagua1', 'delectricidad_med', 'dconexiongas'  son dummies que señalan la presencia de conexión al agua de red, la presencia de un medidor de electricidad y la conexión al gas de red respectivamente."
      ],
      "metadata": {
        "id": "EFDdODVeW44D"
      }
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VL1q9uPd8zK7"
      },
      "source": [
        "El modelo que propondremos consistirá en explicar la condición de vivir en el \"área formal\" a partir de un conjunto de atributos del hogar, incluyendo el ingreso, el tamaño, y un conjunto de variabes indicadores del nivel educativo del jefe de hogar.\n",
        "\n",
        "Proponemos un Modelo Logit. Recordemos que este modelo puede pensar como el modelo de regresión lineal que es transformado mediante la función logística:\n",
        "\n",
        "$$P(F=1)_i=\\frac{1}{1+e^{-Z_i}} $$\n",
        "\n",
        "donde\n",
        "\n",
        "$$Z_i=\\beta_0+\\beta_1IngresoHogar_i+\\beta_2TamañoHogar_i+\\beta_3DSecundarioCompleto_i+\\beta_4DTerciarioCompleto_i+\\epsilon_i$$\n",
        "\n",
        "\n",
        "**Observación:** Una propiedad de este modelo que puede ser útil más adelante es notar que este modelo cumple que el *ratio de las chances* de vivir en el sector formal (definido en este caso como la probabilidad de F=1 sobre la probabilidad de F=0) tomado en logaritmo se asume de forma lineal:\n",
        "\n",
        "\n",
        "$$\\log(\\frac{P(F=1)}{P(F=0)})=\\beta_0+\\beta_1IngresoHogar_i+\\beta_2TamañoHogar_i+\\beta_3DSecundarioCompleto_i+\\beta_4DTerciarioCompleto_i+\\epsilon_i$$\n",
        "\n",
        "Esta última expresión es útil pues, al aparecer P(F=0) explicita que la probabilidad de pertenecer a una categoria se define siempre en relación a otra categoría. Esto será útil para la interpretación del modelo multi-categoría."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MqpJ8QDJ2EPu"
      },
      "source": [
        "# Levantamos los datos\n",
        "datosghp=pd.read_csv('https://raw.githubusercontent.com/rpasquini/urban-econometrics/master/data/formal%20premia%20data%20extract.csv')"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sswLtG3B2ucg",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 487
        },
        "outputId": "6178a652-ac4a-47bd-cb6c-63adaa5028cf"
      },
      "source": [
        "# Cada observación es un hogar\n",
        "datosghp"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "     hora_c  hora_f  bformal  dconurb1  dconurb2  dconurb3  dmore2000  \\\n",
              "0     14.30   14.55        0         0         0         0          1   \n",
              "1     13.30   13.55        0         0         0         0          1   \n",
              "2     14.25   14.50        0         0         0         0          1   \n",
              "3     11.05   11.30        0         0         0         0          1   \n",
              "4      9.35   10.00        0         0         0         0          1   \n",
              "..      ...     ...      ...       ...       ...       ...        ...   \n",
              "545   10.10   10.50        1         0         0         1          0   \n",
              "546   11.40   12.10        1         0         0         1          0   \n",
              "547   12.10   12.35        1         0         0         1          0   \n",
              "548   11.00   11.30        1         0         0         1          0   \n",
              "549    9.30   10.00        1         0         0         1          0   \n",
              "\n",
              "     dprop_soloviv  dprop_ocupant  dinquilinos  ...  m2  dconexionagua1  \\\n",
              "0                1              0            0  ...   2               0   \n",
              "1                0              0            1  ...   2               0   \n",
              "2                1              0            0  ...   2               0   \n",
              "3                1              0            0  ...   2               0   \n",
              "4                1              0            0  ...   1               0   \n",
              "..             ...            ...          ...  ...  ..             ...   \n",
              "545              0              0            0  ...   1               0   \n",
              "546              0              0            1  ...   1               0   \n",
              "547              0              0            0  ...   2               0   \n",
              "548              0              0            0  ...   2               0   \n",
              "549              0              1            0  ...   2               0   \n",
              "\n",
              "     dconexionagua2  dconexionagua3  dpagaelagua  dcloaca  delectricidad_med  \\\n",
              "0                 0               1          0.0        1                  1   \n",
              "1                 0               1          0.0        1                  1   \n",
              "2                 0               1          0.0        0                  0   \n",
              "3                 0               1          NaN        0                  0   \n",
              "4                 0               1          0.0        0                  0   \n",
              "..              ...             ...          ...      ...                ...   \n",
              "545               1               0          0.0        0                  1   \n",
              "546               1               0          0.0        0                  1   \n",
              "547               1               0          0.0        0                  1   \n",
              "548               1               0          0.0        0                  1   \n",
              "549               1               0          0.0        0                  1   \n",
              "\n",
              "     dconexiongas  alquiler2  ingresohogar2  \n",
              "0               0    12000.0        16500.0  \n",
              "1               0     2000.0        12000.0  \n",
              "2               0     3000.0        15000.0  \n",
              "3               0     6000.0         8000.0  \n",
              "4               0        NaN         6800.0  \n",
              "..            ...        ...            ...  \n",
              "545             0     7000.0        25000.0  \n",
              "546             1     5000.0        10000.0  \n",
              "547             1     5000.0        10000.0  \n",
              "548             1        NaN        14000.0  \n",
              "549             0     2000.0        10000.0  \n",
              "\n",
              "[550 rows x 33 columns]"
            ],
            "text/html": [
              "\n",
              "\n",
              "  <div id=\"df-f7f3824e-cf42-4bdf-b9ec-06432b2ecab9\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>hora_c</th>\n",
              "      <th>hora_f</th>\n",
              "      <th>bformal</th>\n",
              "      <th>dconurb1</th>\n",
              "      <th>dconurb2</th>\n",
              "      <th>dconurb3</th>\n",
              "      <th>dmore2000</th>\n",
              "      <th>dprop_soloviv</th>\n",
              "      <th>dprop_ocupant</th>\n",
              "      <th>dinquilinos</th>\n",
              "      <th>...</th>\n",
              "      <th>m2</th>\n",
              "      <th>dconexionagua1</th>\n",
              "      <th>dconexionagua2</th>\n",
              "      <th>dconexionagua3</th>\n",
              "      <th>dpagaelagua</th>\n",
              "      <th>dcloaca</th>\n",
              "      <th>delectricidad_med</th>\n",
              "      <th>dconexiongas</th>\n",
              "      <th>alquiler2</th>\n",
              "      <th>ingresohogar2</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>14.30</td>\n",
              "      <td>14.55</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>12000.0</td>\n",
              "      <td>16500.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>13.30</td>\n",
              "      <td>13.55</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>...</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0.0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>2000.0</td>\n",
              "      <td>12000.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>14.25</td>\n",
              "      <td>14.50</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>3000.0</td>\n",
              "      <td>15000.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>11.05</td>\n",
              "      <td>11.30</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>NaN</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>6000.0</td>\n",
              "      <td>8000.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>9.35</td>\n",
              "      <td>10.00</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>6800.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>545</th>\n",
              "      <td>10.10</td>\n",
              "      <td>10.50</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>7000.0</td>\n",
              "      <td>25000.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>546</th>\n",
              "      <td>11.40</td>\n",
              "      <td>12.10</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>...</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>5000.0</td>\n",
              "      <td>10000.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>547</th>\n",
              "      <td>12.10</td>\n",
              "      <td>12.35</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>5000.0</td>\n",
              "      <td>10000.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>548</th>\n",
              "      <td>11.00</td>\n",
              "      <td>11.30</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>NaN</td>\n",
              "      <td>14000.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>549</th>\n",
              "      <td>9.30</td>\n",
              "      <td>10.00</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>...</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0.0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>2000.0</td>\n",
              "      <td>10000.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>550 rows × 33 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-f7f3824e-cf42-4bdf-b9ec-06432b2ecab9')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "\n",
              "\n",
              "\n",
              "    <div id=\"df-682dd502-bde1-4d11-9c90-2541cf5f8090\">\n",
              "      <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-682dd502-bde1-4d11-9c90-2541cf5f8090')\"\n",
              "              title=\"Suggest charts.\"\n",
              "              style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "      </button>\n",
              "    </div>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "    background-color: #E8F0FE;\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: #1967D2;\n",
              "    height: 32px;\n",
              "    padding: 0 0 0 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: #E2EBFA;\n",
              "    box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: #174EA6;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "    background-color: #3B4455;\n",
              "    fill: #D2E3FC;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart:hover {\n",
              "    background-color: #434B5C;\n",
              "    box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "    filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "    fill: #FFFFFF;\n",
              "  }\n",
              "</style>\n",
              "\n",
              "    <script>\n",
              "      async function quickchart(key) {\n",
              "        const containerElement = document.querySelector('#' + key);\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      }\n",
              "    </script>\n",
              "\n",
              "      <script>\n",
              "\n",
              "function displayQuickchartButton(domScope) {\n",
              "  let quickchartButtonEl =\n",
              "    domScope.querySelector('#df-682dd502-bde1-4d11-9c90-2541cf5f8090 button.colab-df-quickchart');\n",
              "  quickchartButtonEl.style.display =\n",
              "    google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "}\n",
              "\n",
              "        displayQuickchartButton(document);\n",
              "      </script>\n",
              "      <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-f7f3824e-cf42-4bdf-b9ec-06432b2ecab9 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-f7f3824e-cf42-4bdf-b9ec-06432b2ecab9');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n"
            ]
          },
          "metadata": {},
          "execution_count": 3
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fEtbHUPg8HHm"
      },
      "source": [
        "def logit(lista, data):\n",
        "  \"\"\"\n",
        "  Funcion que realiza la regresion Logit\n",
        "  La primera variable de la lista es la dependiente, el resto son las explicativas\n",
        "  data es la base de datos\n",
        "  \"\"\"\n",
        "  datos=data[lista]\n",
        "  # el metodo dropna() me permite eliminar las filas que tienen algun valor missing\n",
        "  datos=datos.dropna()\n",
        "  Y=datos[lista[0]]\n",
        "  X=datos[lista[1:]]\n",
        "  X=sm.add_constant(X)\n",
        "  modelo=sm.Logit(Y,X) # Aqui llamamos al método Logit de StatsModels\n",
        "  modeloajustado=modelo.fit()\n",
        "  print(modeloajustado.summary())\n",
        "  return modeloajustado # Notar que la función devuelve el resultado de ajustar el modelo, este objeto a su vez incluye a una serie de metodos utiles."
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rGBYmgxJ8j5G",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "c16e8950-6da7-48ce-8687-a8cb407c4486"
      },
      "source": [
        "modeloajustado=logit(['bformal', 'ingresohogar2', 'npersonashogar','dedu_seccomp',\t'dedu_ter_o_mas'], datosghp)"
      ],
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Optimization terminated successfully.\n",
            "         Current function value: 0.665327\n",
            "         Iterations 4\n",
            "                           Logit Regression Results                           \n",
            "==============================================================================\n",
            "Dep. Variable:                bformal   No. Observations:                  501\n",
            "Model:                          Logit   Df Residuals:                      496\n",
            "Method:                           MLE   Df Model:                            4\n",
            "Date:                Tue, 18 Jul 2023   Pseudo R-squ.:                 0.03451\n",
            "Time:                        15:26:44   Log-Likelihood:                -333.33\n",
            "converged:                       True   LL-Null:                       -345.24\n",
            "Covariance Type:            nonrobust   LLR p-value:                 8.645e-05\n",
            "==================================================================================\n",
            "                     coef    std err          z      P>|z|      [0.025      0.975]\n",
            "----------------------------------------------------------------------------------\n",
            "const             -0.8477      0.267     -3.176      0.001      -1.371      -0.324\n",
            "ingresohogar2   3.727e-05   1.37e-05      2.713      0.007    1.03e-05    6.42e-05\n",
            "npersonashogar    -0.0141      0.047     -0.299      0.765      -0.106       0.078\n",
            "dedu_seccomp       0.6655      0.211      3.158      0.002       0.252       1.078\n",
            "dedu_ter_o_mas     0.6764      0.371      1.822      0.068      -0.051       1.404\n",
            "==================================================================================\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "MwztcxOnAHkh"
      },
      "source": [
        "¿Cuál es la probabilidad de que una familia de que una familia con cuatro integrantes, ingreso del hogar de $14000, y jefe con secundario incompleto viva en la formalidad? \\\n",
        "A mano:\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "iXUlo8sgAczz",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "a030237c-5d0b-480d-e122-259994f3a52f"
      },
      "source": [
        "import numpy as np\n",
        "\n",
        "z0=-0.8477+0.00003727*14000-0.0141*4+0*0.6655+0*0.67\n",
        "f0=1/(1+np.exp(-z0))\n",
        "f0"
      ],
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.4055674633633202"
            ]
          },
          "metadata": {},
          "execution_count": 6
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nxeefQkrCAPv"
      },
      "source": [
        "Usando el metodo predict. Notar que tenemos que pasar una lista de los valores que tomarán las variables explicativas, en el orden que las incluimos en el modelo. Si bien incluimos 4 variables explicativas, la primera posición que pasamos es un 1 ya que es necesario pasar este valor para indicar la inclusión de la constante ($\\hat{\\beta_0}$) en la predicción.\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "5r-yv44dBRlY",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "961bebfb-8520-41e6-a059-5fc4885c683b"
      },
      "source": [
        "modeloajustado.predict([1,14000,4,0,0])\n"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([0.4056167])"
            ]
          },
          "metadata": {},
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kXDCCtmo-6j4"
      },
      "source": [
        "##  Computando los efectos marginales:\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "StTEGuW-8rFg",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "1ee0dee8-96ed-474a-bf36-60e988f9a9db"
      },
      "source": [
        "# average of the marginal effects in the data (default)\n",
        "probit_margeff = modeloajustado.get_margeff()\n",
        "print('Marginal effects: ')\n",
        "print(probit_margeff.summary())"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Marginal effects: \n",
            "        Logit Marginal Effects       \n",
            "=====================================\n",
            "Dep. Variable:                bformal\n",
            "Method:                          dydx\n",
            "At:                           overall\n",
            "==================================================================================\n",
            "                    dy/dx    std err          z      P>|z|      [0.025      0.975]\n",
            "----------------------------------------------------------------------------------\n",
            "ingresohogar2   8.809e-06   3.16e-06      2.792      0.005    2.62e-06     1.5e-05\n",
            "npersonashogar    -0.0033      0.011     -0.299      0.765      -0.025       0.018\n",
            "dedu_seccomp       0.1573      0.048      3.289      0.001       0.064       0.251\n",
            "dedu_ter_o_mas     0.1599      0.087      1.846      0.065      -0.010       0.330\n",
            "==================================================================================\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EGwd0rYiERbu",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "82a8dd5f-de7e-4c5e-9d12-f89d5892c4bb"
      },
      "source": [
        "# average of the marginal effects in the data (default)\n",
        "probit_margeff = modeloajustado.get_margeff(at='mean')\n",
        "print('Marginal effects: ')\n",
        "print(probit_margeff.summary())"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Marginal effects: \n",
            "        Logit Marginal Effects       \n",
            "=====================================\n",
            "Dep. Variable:                bformal\n",
            "Method:                          dydx\n",
            "At:                              mean\n",
            "==================================================================================\n",
            "                    dy/dx    std err          z      P>|z|      [0.025      0.975]\n",
            "----------------------------------------------------------------------------------\n",
            "ingresohogar2   9.241e-06   3.41e-06      2.712      0.007    2.56e-06    1.59e-05\n",
            "npersonashogar    -0.0035      0.012     -0.299      0.765      -0.026       0.019\n",
            "dedu_seccomp       0.1650      0.052      3.158      0.002       0.063       0.267\n",
            "dedu_ter_o_mas     0.1677      0.092      1.822      0.068      -0.013       0.348\n",
            "==================================================================================\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uQRTWlvXC6I2",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "435effc4-4c34-4f58-c501-c8144b9cff82"
      },
      "source": [
        "probit_margeff = modeloajustado.get_margeff(atexog={1:14000, 2:4, 3:0, 4:0})\n",
        "print('Marginal effects: ')\n",
        "print(probit_margeff.summary())"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Marginal effects: \n",
            "        Logit Marginal Effects       \n",
            "=====================================\n",
            "Dep. Variable:                bformal\n",
            "Method:                          dydx\n",
            "At:                           overall\n",
            "==================================================================================\n",
            "                    dy/dx    std err          z      P>|z|      [0.025      0.975]\n",
            "----------------------------------------------------------------------------------\n",
            "ingresohogar2   8.987e-06   3.35e-06      2.680      0.007    2.41e-06    1.56e-05\n",
            "npersonashogar    -0.0034      0.011     -0.299      0.765      -0.026       0.019\n",
            "dedu_seccomp       0.1604      0.049      3.275      0.001       0.064       0.256\n",
            "dedu_ter_o_mas     0.1631      0.088      1.846      0.065      -0.010       0.336\n",
            "==================================================================================\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bLKrhp18_GUg"
      },
      "source": [
        "# Versión Probit\n",
        "El modelo Probit representa una alternativa al modelo Logit. La función que realiza la transformación en este caso es la [función acumulativa](https://en.wikipedia.org/wiki/Cumulative_distribution_function) de la [Distribución Normal Estandar](https://en.wikipedia.org/wiki/Normal_distribution), que es típicamente denotada $\\Phi$. Es decir:\n",
        "\n",
        "$$P(F=1)=\\Phi(\\beta_0+\\beta_1IngresoHogar_i+\\beta_2TamañoHogar_i+\\beta_3DSecundarioCompleto_i+\\beta_4DTerciarioCompleto_i+\\epsilon_i)$$"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HIplSbtr3MSp"
      },
      "source": [
        "def probit(lista, data):\n",
        "  \"\"\"\n",
        "  Funcion que realiza la regresion probit\n",
        "  La primera variable de la lista es la dependiente, el resto son las explicativas\n",
        "  data es la base de datos\n",
        "  \"\"\"\n",
        "  datos=data[lista]\n",
        "  # el metodo dropna() me permite eliminar las filas que tienen algun valor missing\n",
        "  datos=datos.dropna()\n",
        "  Y=datos[lista[0]]\n",
        "  X=datos[lista[1:]]\n",
        "  X=sm.add_constant(X)\n",
        "  modelo=sm.Probit(Y,X)\n",
        "  modeloajustado=modelo.fit()\n",
        "  print(modeloajustado.summary())\n",
        "  return modeloajustado"
      ],
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "t7VCI9113m4k",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "4ffd27f7-1335-427b-e414-40ae1392b726"
      },
      "source": [
        "modeloajustado=probit(['bformal', 'ingresohogar2', 'npersonashogar','dedu_seccomp',\t'dedu_ter_o_mas'], datosghp)"
      ],
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Optimization terminated successfully.\n",
            "         Current function value: 0.665310\n",
            "         Iterations 4\n",
            "                          Probit Regression Results                           \n",
            "==============================================================================\n",
            "Dep. Variable:                bformal   No. Observations:                  501\n",
            "Model:                         Probit   Df Residuals:                      496\n",
            "Method:                           MLE   Df Model:                            4\n",
            "Date:                Tue, 18 Jul 2023   Pseudo R-squ.:                 0.03453\n",
            "Time:                        15:26:45   Log-Likelihood:                -333.32\n",
            "converged:                       True   LL-Null:                       -345.24\n",
            "Covariance Type:            nonrobust   LLR p-value:                 8.578e-05\n",
            "==================================================================================\n",
            "                     coef    std err          z      P>|z|      [0.025      0.975]\n",
            "----------------------------------------------------------------------------------\n",
            "const             -0.5269      0.164     -3.208      0.001      -0.849      -0.205\n",
            "ingresohogar2   2.306e-05   8.38e-06      2.752      0.006    6.63e-06    3.95e-05\n",
            "npersonashogar    -0.0086      0.029     -0.295      0.768      -0.066       0.048\n",
            "dedu_seccomp       0.4152      0.131      3.166      0.002       0.158       0.672\n",
            "dedu_ter_o_mas     0.4262      0.231      1.849      0.065      -0.026       0.878\n",
            "==================================================================================\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Fycg4Rrj7VJL",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "67bb4ca7-4978-47ef-a202-c3691b5dacfd"
      },
      "source": [
        "probit_margeff = modeloajustado.get_margeff()\n",
        "print('Marginal effects: ')\n",
        "print(probit_margeff.summary())"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Marginal effects: \n",
            "       Probit Marginal Effects       \n",
            "=====================================\n",
            "Dep. Variable:                bformal\n",
            "Method:                          dydx\n",
            "At:                           overall\n",
            "==================================================================================\n",
            "                    dy/dx    std err          z      P>|z|      [0.025      0.975]\n",
            "----------------------------------------------------------------------------------\n",
            "ingresohogar2   8.799e-06   3.12e-06      2.817      0.005    2.68e-06    1.49e-05\n",
            "npersonashogar    -0.0033      0.011     -0.295      0.768      -0.025       0.018\n",
            "dedu_seccomp       0.1584      0.048      3.269      0.001       0.063       0.253\n",
            "dedu_ter_o_mas     0.1626      0.087      1.869      0.062      -0.008       0.333\n",
            "==================================================================================\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "G5PREOazjmUr"
      },
      "source": [
        "# Estimando un Modelo de Localización Residencial en la Ciudad. Aplicación del modelo de múltiples categorías.\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import urllib.request\n",
        "import zipfile\n"
      ],
      "metadata": {
        "id": "U59zRm-Lmm_C"
      },
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# En las ultimas versiones de geopandas, por alguna razon no esta levantando el zip file directamente, entonces voy a implementar la lectura paso a paso\n",
        "#comunas=gpd.GeoDataFrame.from_file(\"https://github.com/rpasquini/econometrics_and_causality/blob/master/comunas.zip?raw=true\")\n",
        "\n",
        "# Specify the URL of the zip file\n",
        "url = \"https://github.com/rpasquini/econometrics_and_causality/blob/master/comunas.zip?raw=true\"\n",
        "\n",
        "# Download the zip file\n",
        "filename, headers = urllib.request.urlretrieve(url)\n",
        "\n",
        "# Extract the contents of the zip file\n",
        "with zipfile.ZipFile(filename, 'r') as zip_ref:\n",
        "    zip_ref.extractall()\n",
        "\n",
        "# Read the shapefile using Geopandas\n",
        "shapefile_path = \"comunas.shp\"\n",
        "comunas = gpd.read_file(shapefile_path)\n"
      ],
      "metadata": {
        "id": "4bSx75FklpJe"
      },
      "execution_count": 15,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "comunas.plot('COMUNAS', legend=True, categorical=True, figsize=(8, 8), cmap='tab20')\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 694
        },
        "id": "ua0OPPEfeTf1",
        "outputId": "a4205859-982a-4fc5-cb47-87f043c8e285"
      },
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<Axes: >"
            ]
          },
          "metadata": {},
          "execution_count": 16
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 800x800 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Estimaremos un modelo de localicalización usando la Encuesta Anual de Hogares de la Dirección General de Estadística y Censos de la Ciudad de Buenos Aires (EAH\n",
        "Véase https://www.estadisticaciudad.gob.ar/eyc/?cat=93)"
      ],
      "metadata": {
        "id": "qAg6Yc3wgre7"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Lo que buscamos es poder explicar/predecir la elección de la comuna donde un hogar vivirá a partir de un conjunto de características explicativas. En este ejemplo,  para simplificar, vamos a trabajar con la encuesta a las personas, y estimaremos el modelo utilizando los datos del jefe del hogar."
      ],
      "metadata": {
        "id": "kuA3IEaqnL6-"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "UI-PPXBEiyO5"
      },
      "source": [
        "dataeah=pd.read_csv(\"https://github.com/rpasquini/econometrics_and_causality/blob/master/eah2016.csv?raw=true\")\n"
      ],
      "execution_count": 26,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AJYnY01qjQ8m",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "76ec8717-404c-40cb-f3f7-4b6dceec0c65"
      },
      "source": [
        "dataeah.columns"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Index(['id', 'nhogar', 'miembro', 'comuna', 'dominio', 'edad', 'sexo',\n",
              "       'parentes_2', 'p5_2', 'p6_a', 'p6_b', 'estado', 'categori', 't13',\n",
              "       't14', 't18', 't28', 't29', 't29a', 't30', 'sem_hs', 't33', 't34',\n",
              "       't35', 't37_cod_2', 't37_coda_2', 't38', 't39', 't40', 't41_cod_2',\n",
              "       't47', 't48', 't51_bis', 'i1', 'i4', 'i6_3', 'i10', 'i11', 'codioph',\n",
              "       'ioph_2', 'codioph_neto', 'ioph_neto_2', 'codios', 'ios_2', 'codioa',\n",
              "       'ioa_2', 'codlab', 'inglab_2', 'codi3_1x', 'i3_1x_2', 'codi3_2x',\n",
              "       'i3_2x_2', 'codi3_8x', 'i3_8x_2', 'codi3_est', 'i3_est_2', 'codi3_otr',\n",
              "       'i3_otr_2', 'codnolab', 'ingnolab_2', 'coding', 'ingtot_2', 'codi_tot',\n",
              "       'itfb_2', 'ipcfb_2', 'e2', 'e4', 'e6', 'e8', 'e12', 'e13', 'e14',\n",
              "       'edad_30', 'nivel', 'aesc', 'm1', 'm1_2', 'm2_anio', 'm3_anio',\n",
              "       'tipcob2_2', 'sn1_1', 'sn1_7', 'sn1_2', 'sn1_3', 'sn1_4', 'sn1_5', 's8',\n",
              "       'sn14', 'sn16', 's28', 's29', 'fexp', 'sexotexto'],\n",
              "      dtype='object')"
            ]
          },
          "metadata": {},
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Para el ejemplo nos quedaremos con las caracteristicas del jefe de hogar, codificado como 1 en la variable parentes_2\n",
        "dataeah=dataeah[dataeah['parentes_2']==1]"
      ],
      "metadata": {
        "id": "ux1IlItFoHM3"
      },
      "execution_count": 28,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Como variables explicativas tomaremos la edad y el monto del ingreso total individual (variable ingtot_2)"
      ],
      "metadata": {
        "id": "H982uvqoo4K4"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nQvI22MSll2I",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 448
        },
        "outputId": "41b82cb9-5916-465d-d9c5-20af9e53a8f8"
      },
      "source": [
        "dataeah[\"ingtot_2\"].hist()"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<Axes: >"
            ]
          },
          "metadata": {},
          "execution_count": 29
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "El modelo a estar sera un Multinomial Logit. Este modelo se puede pensar como una generalización del modelo Logit para una cantidad de K categorías. Podemos pensar a este modelo como la modelización de K-1 ecuaciones tomando como base a una categoría de referencia (la K-esima).  \n",
        "\n",
        "Al igual que el Logit, este modelo tiene la propiedad de que el logaritmo del ratio de las chances entre las categorías se modela de manera lineal. En este caso, el logaritmo del ratio de las chances de una categroria k-esima en relación a la categoría K (la última), se puede escribir como:\n",
        "\n",
        "$$ \\ln(\\frac{\\Pr(Y=k)}{\\Pr(Y=K)})=\\beta_{0,k}+\\beta_{1,k}X_i+\\epsilon_i$$\n",
        "\n",
        "El modelo tiene similares implicancias al Logit de dos categorías: Los efectos del modelo en las probabilidades serán no-lineales (i.e., dependerán de los valores de las variables explicativas) y será necesario estimar los *efectos marginales*.\n",
        "\n"
      ],
      "metadata": {
        "id": "vB6w0iy8pshb"
      }
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Yaow9-oWjtxc"
      },
      "source": [
        "def mnlogit(lista, data):\n",
        "  \"\"\"\n",
        "  Funcion que realiza la regresion Multinomial Logit\n",
        "  La primera variable de la lista es la dependiente, el resto son las explicativas\n",
        "  data es la base de datos\n",
        "  \"\"\"\n",
        "  datos=data[lista]\n",
        "  # el metodo dropna() me permite eliminar las filas que tienen algun valor missing\n",
        "  datos=datos.dropna()\n",
        "  Y=datos[lista[0]]\n",
        "  X=datos[lista[1:]]\n",
        "  X=sm.add_constant(X)\n",
        "  modelo=sm.MNLogit(Y,X)\n",
        "  modeloajustado=modelo.fit()\n",
        "  print(modeloajustado.summary())\n",
        "  return modeloajustado"
      ],
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6DbfWUZdj8xJ",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d3d9197b-993e-49de-fe18-82ddd266e1a9"
      },
      "source": [
        "modeloajustado=mnlogit(['comuna', 'ingtot_2', 'edad'], dataeah)"
      ],
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Optimization terminated successfully.\n",
            "         Current function value: 2.623895\n",
            "         Iterations 6\n",
            "                          MNLogit Regression Results                          \n",
            "==============================================================================\n",
            "Dep. Variable:                 comuna   No. Observations:                 6000\n",
            "Model:                        MNLogit   Df Residuals:                     5958\n",
            "Method:                           MLE   Df Model:                           28\n",
            "Date:                Tue, 18 Jul 2023   Pseudo R-squ.:                 0.01531\n",
            "Time:                        15:27:34   Log-Likelihood:                -15743.\n",
            "converged:                       True   LL-Null:                       -15988.\n",
            "Covariance Type:            nonrobust   LLR p-value:                 9.079e-86\n",
            "==============================================================================\n",
            "  comuna=2       coef    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const         -0.8967      0.198     -4.529      0.000      -1.285      -0.509\n",
            "ingtot_2    3.483e-05   4.51e-06      7.723      0.000     2.6e-05    4.37e-05\n",
            "edad          -0.0014      0.004     -0.402      0.688      -0.008       0.006\n",
            "------------------------------------------------------------------------------\n",
            "  comuna=3       coef    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const         -0.7908      0.196     -4.037      0.000      -1.175      -0.407\n",
            "ingtot_2    7.056e-06   5.25e-06      1.344      0.179   -3.23e-06    1.73e-05\n",
            "edad           0.0084      0.003      2.540      0.011       0.002       0.015\n",
            "------------------------------------------------------------------------------\n",
            "  comuna=4       coef    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const         -0.4191      0.194     -2.155      0.031      -0.800      -0.038\n",
            "ingtot_2    -1.86e-05   5.99e-06     -3.105      0.002   -3.03e-05   -6.86e-06\n",
            "edad           0.0087      0.003      2.689      0.007       0.002       0.015\n",
            "------------------------------------------------------------------------------\n",
            "  comuna=5       coef    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const         -1.3331      0.213     -6.268      0.000      -1.750      -0.916\n",
            "ingtot_2    2.454e-05   4.99e-06      4.922      0.000    1.48e-05    3.43e-05\n",
            "edad           0.0079      0.004      2.153      0.031       0.001       0.015\n",
            "------------------------------------------------------------------------------\n",
            "  comuna=6       coef    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const         -1.6095      0.214     -7.525      0.000      -2.029      -1.190\n",
            "ingtot_2    2.513e-05   4.91e-06      5.116      0.000    1.55e-05    3.48e-05\n",
            "edad           0.0136      0.004      3.783      0.000       0.007       0.021\n",
            "------------------------------------------------------------------------------\n",
            "  comuna=7       coef    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const         -1.2901      0.211     -6.107      0.000      -1.704      -0.876\n",
            "ingtot_2    6.088e-06   5.56e-06      1.096      0.273    -4.8e-06     1.7e-05\n",
            "edad           0.0146      0.003      4.175      0.000       0.008       0.021\n",
            "------------------------------------------------------------------------------\n",
            "  comuna=8       coef    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const          0.0859      0.195      0.441      0.659      -0.296       0.468\n",
            "ingtot_2   -4.074e-05   6.75e-06     -6.034      0.000    -5.4e-05   -2.75e-05\n",
            "edad           0.0028      0.003      0.864      0.387      -0.004       0.009\n",
            "------------------------------------------------------------------------------\n",
            "  comuna=9       coef    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const         -1.3729      0.244     -5.638      0.000      -1.850      -0.896\n",
            "ingtot_2   -2.155e-05   7.59e-06     -2.839      0.005   -3.64e-05   -6.67e-06\n",
            "edad           0.0156      0.004      3.986      0.000       0.008       0.023\n",
            "------------------------------------------------------------------------------\n",
            " comuna=10       coef    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const         -2.0275      0.257     -7.876      0.000      -2.532      -1.523\n",
            "ingtot_2    5.732e-06   6.58e-06      0.871      0.384   -7.16e-06    1.86e-05\n",
            "edad           0.0186      0.004      4.477      0.000       0.010       0.027\n",
            "------------------------------------------------------------------------------\n",
            " comuna=11       coef    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const         -2.4954      0.294     -8.489      0.000      -3.072      -1.919\n",
            "ingtot_2    1.422e-05   6.88e-06      2.069      0.039    7.47e-07    2.77e-05\n",
            "edad           0.0184      0.005      3.866      0.000       0.009       0.028\n",
            "------------------------------------------------------------------------------\n",
            " comuna=12       coef    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const         -1.3886      0.221     -6.292      0.000      -1.821      -0.956\n",
            "ingtot_2    1.829e-05   5.37e-06      3.408      0.001    7.77e-06    2.88e-05\n",
            "edad           0.0091      0.004      2.435      0.015       0.002       0.016\n",
            "------------------------------------------------------------------------------\n",
            " comuna=13       coef    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const         -1.8786      0.217     -8.664      0.000      -2.304      -1.454\n",
            "ingtot_2    3.474e-05   4.59e-06      7.565      0.000    2.57e-05    4.37e-05\n",
            "edad           0.0145      0.004      3.961      0.000       0.007       0.022\n",
            "------------------------------------------------------------------------------\n",
            " comuna=14       coef    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const         -1.5506      0.211     -7.353      0.000      -1.964      -1.137\n",
            "ingtot_2    4.143e-05   4.43e-06      9.350      0.000    3.27e-05    5.01e-05\n",
            "edad           0.0060      0.004      1.612      0.107      -0.001       0.013\n",
            "------------------------------------------------------------------------------\n",
            " comuna=15       coef    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const         -1.5228      0.225     -6.766      0.000      -1.964      -1.082\n",
            "ingtot_2    1.733e-05   5.46e-06      3.174      0.002    6.63e-06     2.8e-05\n",
            "edad           0.0112      0.004      2.975      0.003       0.004       0.019\n",
            "==============================================================================\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "probit_margeff = modeloajustado.get_margeff(atexog={1:50000, 2:30})\n",
        "print('Marginal effects: ')\n",
        "print(probit_margeff.summary())"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8x1-odrzvACi",
        "outputId": "4284c0db-95f6-4905-938c-09c506cce96c"
      },
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Marginal effects: \n",
            "       MNLogit Marginal Effects      \n",
            "=====================================\n",
            "Dep. Variable:                 comuna\n",
            "Method:                          dydx\n",
            "At:                           overall\n",
            "==============================================================================\n",
            "  comuna=1      dy/dx    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "ingtot_2   -1.723e-06   1.28e-07    -13.476      0.000   -1.97e-06   -1.47e-06\n",
            "edad          -0.0006      0.000     -2.842      0.004      -0.001      -0.000\n",
            "------------------------------------------------------------------------------\n",
            "  comuna=2      dy/dx    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "ingtot_2    2.149e-06    5.1e-07      4.217      0.000    1.15e-06    3.15e-06\n",
            "edad          -0.0016      0.000     -3.200      0.001      -0.003      -0.001\n",
            "------------------------------------------------------------------------------\n",
            "  comuna=3      dy/dx    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "ingtot_2   -9.791e-07   1.44e-07     -6.808      0.000   -1.26e-06   -6.97e-07\n",
            "edad        4.839e-05      0.000      0.306      0.760      -0.000       0.000\n",
            "------------------------------------------------------------------------------\n",
            "  comuna=4      dy/dx    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "ingtot_2   -1.065e-06   9.43e-08    -11.297      0.000   -1.25e-06    -8.8e-07\n",
            "edad        2.787e-05   6.47e-05      0.431      0.667   -9.89e-05       0.000\n",
            "------------------------------------------------------------------------------\n",
            "  comuna=5      dy/dx    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "ingtot_2    1.936e-07   3.03e-07      0.639      0.523      -4e-07    7.87e-07\n",
            "edad        1.742e-05      0.000      0.071      0.943      -0.000       0.000\n",
            "------------------------------------------------------------------------------\n",
            "  comuna=6      dy/dx    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "ingtot_2    2.278e-07   2.78e-07      0.818      0.413   -3.18e-07    7.73e-07\n",
            "edad           0.0005      0.000      2.383      0.017     8.6e-05       0.001\n",
            "------------------------------------------------------------------------------\n",
            "  comuna=7      dy/dx    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "ingtot_2    -7.24e-07   1.08e-07     -6.724      0.000   -9.35e-07   -5.13e-07\n",
            "edad           0.0003      0.000      2.680      0.007    8.26e-05       0.001\n",
            "------------------------------------------------------------------------------\n",
            "  comuna=8      dy/dx    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "ingtot_2   -7.535e-07    1.1e-07     -6.845      0.000   -9.69e-07   -5.38e-07\n",
            "edad       -5.745e-05   3.62e-05     -1.585      0.113      -0.000    1.36e-05\n",
            "------------------------------------------------------------------------------\n",
            "  comuna=9      dy/dx    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "ingtot_2   -4.666e-07   6.38e-08     -7.312      0.000   -5.92e-07   -3.42e-07\n",
            "edad        8.465e-05   3.74e-05      2.265      0.024    1.14e-05       0.000\n",
            "------------------------------------------------------------------------------\n",
            " comuna=10      dy/dx    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "ingtot_2    -3.92e-07   7.28e-08     -5.385      0.000   -5.35e-07   -2.49e-07\n",
            "edad           0.0003   8.21e-05      3.137      0.002    9.66e-05       0.000\n",
            "------------------------------------------------------------------------------\n",
            " comuna=11      dy/dx    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "ingtot_2   -1.822e-07   9.99e-08     -1.825      0.068   -3.78e-07    1.35e-08\n",
            "edad           0.0002   8.75e-05      2.748      0.006     6.9e-05       0.000\n",
            "------------------------------------------------------------------------------\n",
            " comuna=12      dy/dx    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "ingtot_2   -2.553e-07   2.15e-07     -1.185      0.236   -6.78e-07    1.67e-07\n",
            "edad        9.198e-05      0.000      0.502      0.616      -0.000       0.000\n",
            "------------------------------------------------------------------------------\n",
            " comuna=13      dy/dx    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "ingtot_2    1.284e-06   3.57e-07      3.595      0.000    5.84e-07    1.98e-06\n",
            "edad           0.0007      0.000      2.826      0.005       0.000       0.001\n",
            "------------------------------------------------------------------------------\n",
            " comuna=14      dy/dx    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "ingtot_2    2.966e-06   5.02e-07      5.905      0.000    1.98e-06    3.95e-06\n",
            "edad          -0.0003      0.000     -0.624      0.532      -0.001       0.001\n",
            "------------------------------------------------------------------------------\n",
            " comuna=15      dy/dx    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "ingtot_2   -2.807e-07   1.92e-07     -1.459      0.145   -6.58e-07    9.65e-08\n",
            "edad           0.0002      0.000      1.254      0.210      -0.000       0.001\n",
            "==============================================================================\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "_xDvr4GYkZwC",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "24e8c46f-5230-40bd-bf7d-c3d54db36ade"
      },
      "source": [
        "modeloajustado.predict([1,20000,60])"
      ],
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[0.10010879, 0.07522822, 0.08659332, 0.07660997, 0.06906744,\n",
              "        0.07482682, 0.07454623, 0.05729652, 0.04206905, 0.04501185,\n",
              "        0.03300705, 0.0621908 , 0.0732559 , 0.06954885, 0.06063918]])"
            ]
          },
          "metadata": {},
          "execution_count": 33
        }
      ]
    }
  ]
}