notebooks/Aula_07.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ystYaKIiY2wS"
      },
      "source": [
        "# Aula 7 - Modelo Linear Genérico, codificação de variáveis e moderação/mediação"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "D67fj2XRY2wa"
      },
      "source": [
        "## Modelo Linear Genérico (GLM)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lLG2Yltn9yoM"
      },
      "source": [
        "Estrutura matemática utilizada em muitas análises estatísticas, incluindo Regressão Múltipla e ANOVA (normalmente representada como uma análise diferente, embora - na realidade - seja regressão múltipla).\n",
        "\n",
        "Características:\n",
        "\n",
        "**Linear**: pares de variáveis que (acredita-se) possuirem uma relação linear.\n",
        "**Aditivo**: se um conjunto de variáveis prevê outra variável, os efeitos são considerados como aditivos.\n",
        "\n",
        "MAS, isso *não elimina a necessidade do teste de efeitos causados por não linearidade ou não aditividade*. \n",
        "\n",
        "Permite a configuração de diversos testes:\n",
        "\n",
        "**Transformação de Variáveis**: transformar variáveis não lineares para lineares\n",
        "**Análises de moderação**: *simulação de um falso GLM* para testar efeitos não aditivos.\n",
        "\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "a2ZkRCYB-jGk"
      },
      "source": [
        "Modelo de Regressão Linear simples = Modelo de one-way ANOVA:\n",
        "**Ŷ = B0 + B1X + Ɛ (erro aleatório)**\n",
        "\n",
        "- Ŷ = valor previsto para a variável de resposta Y\n",
        "- B0 = interseção (previsão do valor de Y quando X = 0)\n",
        "- B1 = coeficiente de regressão (inclinação da reta);\n",
        "- X  = variável de previsão (nominal em anova)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ei8OH4lBG-Jc"
      },
      "source": [
        "Modelo de Regressão Múltipla = Modelo de ANOVA fatorial:\n",
        "\n",
        "Ŷ = B0 + B1X1 + B2X2 + B3X3 + ... + BkXk  + Ɛ \n",
        "\n",
        "**Ŷ = B0 + ∑(BkXk)**\n",
        "\n",
        "- Ŷ = valor previsto para a variável de resposta Y\n",
        "- B0 = previsão do valor de Y quando X = 0\n",
        "- Xk  = variáveis de previsão (nominais em ANOVA fatorial)\n",
        "- Bk = coeficientes de regressão não padronizados\n",
        "- Y – Ŷ = residuais (erro de predição)\n",
        "- K = número de variáveis de previsão"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vlBVCeVJ4OAf"
      },
      "source": [
        "Análise de Variância - ANOVA:\n",
        "Adequado quando os preditores (variáveis independentes) são **categóricas** e o resultado (variável dependente) é **contínua**.\n",
        "\n",
        "A aplicação mais comum é para a análise de dados de experimentos randomizados, principalmente aqueles que geram mais de duas médias.\n",
        "\n",
        "Quando somente 2 médias são geradas, utiliza-se testes-t.\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "w6nUyEYSY2wb"
      },
      "outputs": [],
      "source": [
        "# Importar bibliotecas\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "import seaborn as sns\n",
        "import statsmodels.formula.api as smf\n",
        "from scipy.stats import linregress"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PTFWekrn40Ca"
      },
      "source": [
        "## Codificação de variáveis"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2rf3leBI48uI"
      },
      "source": [
        "Sistema de codificação de **preditores categoricos** em análises de regressão.\n",
        "**Binário**: pode-se codificar diretamente na coluna como '0' (categoria padrão) e '1' (categoria de comparação).\n",
        "**Três categorias ou mais**: pode-se criar uma coluna para cada categoria (exceto uma, que será a padrão) e codificar de algumas maneiras:\n",
        "- **Codificação fictícia**: '1' (efeito presente) ou '0' (efeito ausente).\n",
        "- **Codificação de efeito sem pesos**: '-1' para a categoria padrão e '1' (efeito presente) ou '0' (efeito ausente). Esta codificação 'zera' cada coluna resultando no efeito de cada categoria sem o efeito da categoria padrão, mas não leva em consideração a diferença de amostragem de cada categoria. Bom quando todas as categorias tem o mesmo peso (n de observações).\n",
        "- **Codificação de efeitos com pesos**: \n",
        "-- Categoria padrão: -Ncat da coluna/N da categoria padrão.\n",
        "-- Para cada categoria: Ncat/Ncat padrão (quando presente) e '0' quando ausente."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "eG-oRisI71ec"
      },
      "source": [
        "## Moderação "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "V6f0cpC1_nKh"
      },
      "source": [
        "Podemos dizer que uma variável 'X' é moderada por uma segunda variável 'Z' quando a **inclusão da interação entre as duas resulta em uma melhora no modelo de regressão**. Isso acontece quando a relação entre as variáveis X e Y varia como uma função de Z, ou seja, **não é linear ao longo de 'Z'**.\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "pQEDVRoq8NFy"
      },
      "outputs": [],
      "source": [
        "# Exemplo sem interação (linear ao longo de Z)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "GMliBPn78RuO"
      },
      "outputs": [],
      "source": [
        "# Exemplo com interação (não linear ao longo de Z)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "EaIszQNVjJNi"
      },
      "source": [
        "# Atividade"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UN_SpXV3A4XD"
      },
      "source": [
        "### Parte 01: Análise de Moderação"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "id": "4XlqE5CYY2wd"
      },
      "outputs": [],
      "source": [
        "# Ler base de dados\n",
        "threats = pd.read_excel('../data/iq_threats.xlsx')"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "GtQ2MSaCBBg6"
      },
      "source": [
        "Uma pesquisa experimental investiga os efeitos de ameaças de estereotipo em testes de inteligência.\n",
        "- A variável dependente (Y) é o escore no teste de inteligência.\n",
        "- A variável independente é a condição de tratamento (3 níveis: controle, ameaça explícita e ameaça implícita).\n",
        "- A variável de Moderação é o escore em tarefa de memória de trabalho.\n",
        "\n",
        "Tamanho da amostra N=150 (n=50).\n",
        "\n",
        "\\** a condição experimental é categórica, e já possui uma codificação fictícia: \n",
        "- Grupo de controle = referência\n",
        "- D1 = Ameaça Explícita\n",
        "- D2 = Ameaça Implícita"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gOW6qPjGL4VV",
        "outputId": "10f3ac05-fce8-439b-97ff-9072c0ec4fff"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 150 entries, 0 to 149\n",
            "Data columns (total 7 columns):\n",
            " #   Column       Non-Null Count  Dtype  \n",
            "---  ------       --------------  -----  \n",
            " 0   subject      150 non-null    int64  \n",
            " 1   condition    150 non-null    object \n",
            " 2   IQ           150 non-null    int64  \n",
            " 3   WM           150 non-null    int64  \n",
            " 4   WM.centered  150 non-null    float64\n",
            " 5   D1           150 non-null    int64  \n",
            " 6   D2           150 non-null    int64  \n",
            "dtypes: float64(1), int64(5), object(1)\n",
            "memory usage: 8.3+ KB\n"
          ]
        }
      ],
      "source": [
        "# Verifique as colunas dos dados\n",
        "threats.info()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "id": "4V0aGF6iCTMT",
        "outputId": "80be310a-89d2-4667-9cea-ffc889c38fdd"
      },
      "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>subject</th>\n",
              "      <th>IQ</th>\n",
              "      <th>WM</th>\n",
              "      <th>WM.centered</th>\n",
              "      <th>D1</th>\n",
              "      <th>D2</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>150.00</td>\n",
              "      <td>150.00</td>\n",
              "      <td>150.00</td>\n",
              "      <td>150.00</td>\n",
              "      <td>150.00</td>\n",
              "      <td>150.00</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>75.50</td>\n",
              "      <td>66.02</td>\n",
              "      <td>99.08</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.33</td>\n",
              "      <td>0.33</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>43.45</td>\n",
              "      <td>27.78</td>\n",
              "      <td>18.37</td>\n",
              "      <td>18.37</td>\n",
              "      <td>0.47</td>\n",
              "      <td>0.47</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>1.00</td>\n",
              "      <td>21.00</td>\n",
              "      <td>55.00</td>\n",
              "      <td>-44.08</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.00</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>38.25</td>\n",
              "      <td>44.25</td>\n",
              "      <td>86.00</td>\n",
              "      <td>-13.08</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.00</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>75.50</td>\n",
              "      <td>58.50</td>\n",
              "      <td>96.50</td>\n",
              "      <td>-2.58</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.00</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>112.75</td>\n",
              "      <td>81.00</td>\n",
              "      <td>111.00</td>\n",
              "      <td>11.92</td>\n",
              "      <td>1.00</td>\n",
              "      <td>1.00</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>150.00</td>\n",
              "      <td>141.00</td>\n",
              "      <td>159.00</td>\n",
              "      <td>59.92</td>\n",
              "      <td>1.00</td>\n",
              "      <td>1.00</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       subject      IQ      WM  WM.centered      D1      D2\n",
              "count   150.00  150.00  150.00       150.00  150.00  150.00\n",
              "mean     75.50   66.02   99.08         0.00    0.33    0.33\n",
              "std      43.45   27.78   18.37        18.37    0.47    0.47\n",
              "min       1.00   21.00   55.00       -44.08    0.00    0.00\n",
              "25%      38.25   44.25   86.00       -13.08    0.00    0.00\n",
              "50%      75.50   58.50   96.50        -2.58    0.00    0.00\n",
              "75%     112.75   81.00  111.00        11.92    1.00    1.00\n",
              "max     150.00  141.00  159.00        59.92    1.00    1.00"
            ]
          },
          "execution_count": 11,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "#Vizualize as estatísticas de sumário \n",
        "threats.describe().round(2)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 237
        },
        "id": "rqgY8mHKN3fK",
        "outputId": "55fa0fdb-9d84-4b89-e3fb-b05f44b8be22"
      },
      "outputs": [
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/tmp/ipykernel_52752/2306592302.py:2: FutureWarning: The default value of numeric_only in DataFrame.corr is deprecated. In a future version, it will default to False. Select only valid columns or specify the value of numeric_only to silence this warning.\n",
            "  cor = threats.corr().round(2)\n"
          ]
        },
        {
          "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>subject</th>\n",
              "      <th>IQ</th>\n",
              "      <th>WM</th>\n",
              "      <th>WM.centered</th>\n",
              "      <th>D1</th>\n",
              "      <th>D2</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>subject</th>\n",
              "      <td>1.00</td>\n",
              "      <td>-0.71</td>\n",
              "      <td>-0.16</td>\n",
              "      <td>-0.16</td>\n",
              "      <td>-0.00</td>\n",
              "      <td>0.82</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>IQ</th>\n",
              "      <td>-0.71</td>\n",
              "      <td>1.00</td>\n",
              "      <td>0.34</td>\n",
              "      <td>0.34</td>\n",
              "      <td>-0.35</td>\n",
              "      <td>-0.46</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>WM</th>\n",
              "      <td>-0.16</td>\n",
              "      <td>0.34</td>\n",
              "      <td>1.00</td>\n",
              "      <td>1.00</td>\n",
              "      <td>0.07</td>\n",
              "      <td>-0.19</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>WM.centered</th>\n",
              "      <td>-0.16</td>\n",
              "      <td>0.34</td>\n",
              "      <td>1.00</td>\n",
              "      <td>1.00</td>\n",
              "      <td>0.07</td>\n",
              "      <td>-0.19</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>D1</th>\n",
              "      <td>-0.00</td>\n",
              "      <td>-0.35</td>\n",
              "      <td>0.07</td>\n",
              "      <td>0.07</td>\n",
              "      <td>1.00</td>\n",
              "      <td>-0.50</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>D2</th>\n",
              "      <td>0.82</td>\n",
              "      <td>-0.46</td>\n",
              "      <td>-0.19</td>\n",
              "      <td>-0.19</td>\n",
              "      <td>-0.50</td>\n",
              "      <td>1.00</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "             subject    IQ    WM  WM.centered    D1    D2\n",
              "subject         1.00 -0.71 -0.16        -0.16 -0.00  0.82\n",
              "IQ             -0.71  1.00  0.34         0.34 -0.35 -0.46\n",
              "WM             -0.16  0.34  1.00         1.00  0.07 -0.19\n",
              "WM.centered    -0.16  0.34  1.00         1.00  0.07 -0.19\n",
              "D1             -0.00 -0.35  0.07         0.07  1.00 -0.50\n",
              "D2              0.82 -0.46 -0.19        -0.19 -0.50  1.00"
            ]
          },
          "execution_count": 13,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Verifique a correlação entre as Variáveis\n",
        "cor = threats.corr().round(2)\n",
        "cor"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "pAtrnJY_FaGo"
      },
      "source": [
        "Há um efeito de Ameaça de estereótipo? \n",
        "\n",
        "Teste fazendo um modelo\n",
        "- Regressão linear Múltipla prevendo IQ à partir das condições de ameaça D1 e D2 (QI / D1+D2)."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tc59KPMo1iY0",
        "outputId": "8d7b307f-9cf4-4673-e598-fe1ebd69750e"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Intercept    97.88\n",
            "D1          -45.72\n",
            "D2          -49.86\n",
            "dtype: float64\n",
            "                            OLS Regression Results                            \n",
            "==============================================================================\n",
            "Dep. Variable:                     IQ   R-squared:                       0.666\n",
            "Model:                            OLS   Adj. R-squared:                  0.661\n",
            "Method:                 Least Squares   F-statistic:                     146.6\n",
            "Date:                Mon, 31 Oct 2022   Prob (F-statistic):           9.88e-36\n",
            "Time:                        20:12:34   Log-Likelihood:                -628.72\n",
            "No. Observations:                 150   AIC:                             1263.\n",
            "Df Residuals:                     147   BIC:                             1272.\n",
            "Df Model:                           2                                         \n",
            "Covariance Type:            nonrobust                                         \n",
            "==============================================================================\n",
            "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "Intercept     97.8800      2.286     42.826      0.000      93.363     102.397\n",
            "D1           -45.7200      3.232    -14.145      0.000     -52.108     -39.332\n",
            "D2           -49.8600      3.232    -15.426      0.000     -56.248     -43.472\n",
            "==============================================================================\n",
            "Omnibus:                        1.182   Durbin-Watson:                   1.595\n",
            "Prob(Omnibus):                  0.554   Jarque-Bera (JB):                0.785\n",
            "Skew:                           0.126   Prob(JB):                        0.676\n",
            "Kurtosis:                       3.249   Cond. No.                         3.73\n",
            "==============================================================================\n",
            "\n",
            "Notes:\n",
            "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
          ]
        }
      ],
      "source": [
        "# Regressão Linear Múltipla IQ / D1 + D2:\n",
        "IQ_S = smf.ols('IQ ~ D1 + D2', data=threats).fit()\n",
        "print(IQ_S.params)\n",
        "print(IQ_S.summary())"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MAEDPQ1i3PzZ",
        "outputId": "b6c9c21d-d8cb-44e0-db45-d6a1a2157f5e"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Intercept    97.88\n",
            "D1          -45.72\n",
            "D2          -49.86\n",
            "dtype: float64\n",
            "                 Generalized Linear Model Regression Results                  \n",
            "==============================================================================\n",
            "Dep. Variable:                     IQ   No. Observations:                  150\n",
            "Model:                            GLM   Df Residuals:                      147\n",
            "Model Family:                Gaussian   Df Model:                            2\n",
            "Link Function:               identity   Scale:                          261.18\n",
            "Method:                          IRLS   Log-Likelihood:                -628.72\n",
            "Date:                Mon, 31 Oct 2022   Deviance:                       38393.\n",
            "Time:                        20:12:45   Pearson chi2:                 3.84e+04\n",
            "No. Iterations:                     3   Pseudo R-squ. (CS):             0.8583\n",
            "Covariance Type:            nonrobust                                         \n",
            "==============================================================================\n",
            "                 coef    std err          z      P>|z|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "Intercept     97.8800      2.286     42.826      0.000      93.400     102.360\n",
            "D1           -45.7200      3.232    -14.145      0.000     -52.055     -39.385\n",
            "D2           -49.8600      3.232    -15.426      0.000     -56.195     -43.525\n",
            "==============================================================================\n"
          ]
        }
      ],
      "source": [
        "# Regressão Linear Múltipla IQ / D1 + D2:\n",
        "IQ_S = smf.glm('IQ ~ D1 + D2', data=threats).fit()\n",
        "print(IQ_S.params)\n",
        "print(IQ_S.summary())"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "04UUDQ3SCq4V"
      },
      "source": [
        "RESPOSTA (interpretação das análises acima): \n",
        "\n",
        "Sim, há efeito da ameaça de estereótipo. Isso pode ser visto nos coeficientes de regressão das duas condições de ameaça (ambos significativos) e no modelo (R2=0.67) e significância do mesmo (p<.01)."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZQ2QzlWlDYMG"
      },
      "source": [
        "Há efeito de moderação da Memória de trabalho no tipo de ameaça?\n",
        "\n",
        "Teste (Modelo_1) para efeitos primários das variáves:\n",
        "- Regressão Linear Múltipla de IQ / WM+D1+D2.\n",
        "\n",
        "Teste (Modelo 2) para efeitos de moderação:\n",
        "- Regressão Linear Múltipla de IQ / WM+D1+D2+**WM\\*D1+WM\\*D2**."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ttoNab6nCt51",
        "outputId": "89091683-d30e-471c-ff68-086720eb8bae"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Intercept    59.786350\n",
            "WM            0.372809\n",
            "D1          -45.205523\n",
            "D2          -46.907351\n",
            "dtype: float64\n",
            "                            OLS Regression Results                            \n",
            "==============================================================================\n",
            "Dep. Variable:                     IQ   R-squared:                       0.725\n",
            "Model:                            OLS   Adj. R-squared:                  0.719\n",
            "Method:                 Least Squares   F-statistic:                     128.1\n",
            "Date:                Mon, 31 Oct 2022   Prob (F-statistic):           1.08e-40\n",
            "Time:                        20:13:05   Log-Likelihood:                -614.24\n",
            "No. Observations:                 150   AIC:                             1236.\n",
            "Df Residuals:                     146   BIC:                             1249.\n",
            "Df Model:                           3                                         \n",
            "Covariance Type:            nonrobust                                         \n",
            "==============================================================================\n",
            "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "Intercept     59.7864      7.144      8.369      0.000      45.668      73.905\n",
            "WM             0.3728      0.067      5.575      0.000       0.241       0.505\n",
            "D1           -45.2055      2.946    -15.343      0.000     -51.029     -39.382\n",
            "D2           -46.9074      2.992    -15.677      0.000     -52.821     -40.994\n",
            "==============================================================================\n",
            "Omnibus:                        3.946   Durbin-Watson:                   1.444\n",
            "Prob(Omnibus):                  0.139   Jarque-Bera (JB):                4.598\n",
            "Skew:                          -0.108   Prob(JB):                        0.100\n",
            "Kurtosis:                       3.830   Cond. No.                         611.\n",
            "==============================================================================\n",
            "\n",
            "Notes:\n",
            "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
          ]
        }
      ],
      "source": [
        "# Teste (Modelo_1) para efeitos primários das variáveis\n",
        "Modelo_1 = smf.ols('IQ ~ WM + D1 + D2', threats).fit()\n",
        "print(Modelo_1.params)\n",
        "print(Modelo_1.summary())"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 17,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Fo-3KslRE27O",
        "outputId": "152461d6-40c6-43bc-cfa7-993dd126185a"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Intercept    85.585112\n",
            "WM            0.120326\n",
            "D1          -93.095177\n",
            "D2          -79.897025\n",
            "WM:D1         0.471639\n",
            "WM:D2         0.328772\n",
            "dtype: float64\n",
            "                            OLS Regression Results                            \n",
            "==============================================================================\n",
            "Dep. Variable:                     IQ   R-squared:                       0.741\n",
            "Model:                            OLS   Adj. R-squared:                  0.732\n",
            "Method:                 Least Squares   F-statistic:                     82.35\n",
            "Date:                Mon, 31 Oct 2022   Prob (F-statistic):           1.78e-40\n",
            "Time:                        20:13:20   Log-Likelihood:                -609.67\n",
            "No. Observations:                 150   AIC:                             1231.\n",
            "Df Residuals:                     144   BIC:                             1249.\n",
            "Df Model:                           5                                         \n",
            "Covariance Type:            nonrobust                                         \n",
            "==============================================================================\n",
            "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "Intercept     85.5851     11.358      7.535      0.000      63.136     108.034\n",
            "WM             0.1203      0.109      1.100      0.273      -0.096       0.336\n",
            "D1           -93.0952     16.857     -5.523      0.000    -126.415     -59.775\n",
            "D2           -79.8970     15.477     -5.162      0.000    -110.489     -49.305\n",
            "WM:D1          0.4716      0.164      2.880      0.005       0.148       0.795\n",
            "WM:D2          0.3288      0.155      2.125      0.035       0.023       0.635\n",
            "==============================================================================\n",
            "Omnibus:                        5.452   Durbin-Watson:                   1.547\n",
            "Prob(Omnibus):                  0.065   Jarque-Bera (JB):                8.199\n",
            "Skew:                           0.001   Prob(JB):                       0.0166\n",
            "Kurtosis:                       4.145   Cond. No.                     2.10e+03\n",
            "==============================================================================\n",
            "\n",
            "Notes:\n",
            "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
            "[2] The condition number is large, 2.1e+03. This might indicate that there are\n",
            "strong multicollinearity or other numerical problems.\n"
          ]
        }
      ],
      "source": [
        "# Teste (Modelo_2) para efeitos de moderação\n",
        "Modelo_2 = smf.ols('IQ ~ WM + D1 + D2 + WM*D1 + WM*D2', threats).fit()\n",
        "print(Modelo_2.params)\n",
        "print(Modelo_2.summary())"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KtfaHqajBc5U",
        "outputId": "e41687d4-b565-40d9-e5a1-dc7878a0c64c"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "   df_resid           ssr  df_diff      ss_diff         F    Pr(>F)\n",
            "0     146.0  31655.041707      0.0          NaN       NaN       NaN\n",
            "1     144.0  29783.712093      2.0  1871.329614  4.523806  0.012433\n"
          ]
        }
      ],
      "source": [
        "# Comparando os modelos 1 e 2 acima:\n",
        "from statsmodels.stats.api import anova_lm\n",
        "\n",
        "mod_compare = anova_lm(Modelo_1, Modelo_2)\n",
        "print(mod_compare)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 581
        },
        "id": "eL6hrMnwIvsm",
        "outputId": "9ef7b10d-3be4-4454-c572-c5c801505a80"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 901.25x800 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Gráfico de Dispersão por condição\n",
        "graph = sns.lmplot( data=threats, x='WM', y='IQ', hue='condition', height=8)\n",
        "# Nomeação dos eixos\n",
        "graph.set_axis_labels('Working Memory Score', 'IQ Score')\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "aDywIRgxEVyQ"
      },
      "source": [
        "Com base na interpretação dos modelos acima, responda: \n",
        "\n",
        "- 1. O 'Modelo_2' desempenha melhor que o 'Modelo_1'? Elabore.\n",
        "\n",
        "- 2. O coeficiente de regressão das variáveis de moderação (produtos) são significativos? O que isso significa? Elabore.\n",
        "\n",
        "- 3. Há moderação da ameaça? Elabore."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nzNHpL0v9nyg"
      },
      "source": [
        "R1. Em valores absolutos sim, embora ambos os modelos sejam bem similares. O modelo 1 apresenta todos os coeficientes de regressão com validade estatística, enquanto o modelo 2 tem o coeficiente de regressão de 'Working Memory' não significativo, mas as interações são significativas, indicando moderação da variável em questão.\n",
        "\n",
        "R2. Sim. Que a variável 'Ameaça (Threat)' não varia de maneira linear ao longo de 'WM'.\n",
        "\n",
        "R3. Sim. O modelo melhora com a inclusão da interação entre variáveis. Isso pode ser confirmado na comparação de modelos que mostra uma diferença estatisticamente significante (p<.01)."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NQH1jLjpBXib"
      },
      "source": [
        "### Parte 02: Análise de Mediação"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-j192erzBiIc"
      },
      "source": [
        "Uma pesquisa experimental investiga os efeitos de ameaças de estereotipo em testes de inteligência.\n",
        "- A variável dependente (Y) é o escore no teste de inteligência.\n",
        "- A variável independente (X) é a condição de tratamento (2 níveis: controle, ameaça)\n",
        "- A variável de mediação (M) é o escore em tarefa de memória de trabalho.\n",
        "\n",
        "Tamanho da amostra N=100 (n=50)."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "y7GCCYUtARWv"
      },
      "outputs": [],
      "source": [
        "# Ler base de dados\n",
        "Threat = pd.read_excel('/content/drive/MyDrive/!UFSM/Analise_Estatistica/dataset/IQ_Threat_WM.xlsx')"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "47w7RL5pFJX2",
        "outputId": "b7cd6d01-968d-42ae-cffb-566fd147fd38"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 100 entries, 0 to 99\n",
            "Data columns (total 4 columns):\n",
            " #   Column     Non-Null Count  Dtype \n",
            "---  ------     --------------  ----- \n",
            " 0   subject    100 non-null    int64 \n",
            " 1   condition  100 non-null    object\n",
            " 2   IQ         100 non-null    int64 \n",
            " 3   WM         100 non-null    int64 \n",
            "dtypes: int64(3), object(1)\n",
            "memory usage: 3.2+ KB\n"
          ]
        }
      ],
      "source": [
        "# Vizualize as estatísticas de sumário (por condição de tratamento)\n",
        "Threat.info()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xGGqPppyFJul"
      },
      "source": [
        "Analise o efeito de mediação através de 3 modelos de regressão linear:\n",
        "- Mod_01 (Y~X) - IQ / Condição\n",
        "- Mod_02 (Y~X+M) - IQ / Condição + WM\n",
        "- Mod_03 (M~X) - WM / condição"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NMHtEN8_GVvg",
        "outputId": "0c98bf5b-79fd-4840-8ba5-8fac4f879aff"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Intercept              97.32\n",
            "condition[T.threat]   -11.00\n",
            "dtype: float64\n",
            "                            OLS Regression Results                            \n",
            "==============================================================================\n",
            "Dep. Variable:                     IQ   R-squared:                       0.126\n",
            "Model:                            OLS   Adj. R-squared:                  0.117\n",
            "Method:                 Least Squares   F-statistic:                     14.11\n",
            "Date:                Fri, 28 Oct 2022   Prob (F-statistic):           0.000293\n",
            "Time:                        23:55:06   Log-Likelihood:                -409.27\n",
            "No. Observations:                 100   AIC:                             822.5\n",
            "Df Residuals:                      98   BIC:                             827.8\n",
            "Df Model:                           1                                         \n",
            "Covariance Type:            nonrobust                                         \n",
            "=======================================================================================\n",
            "                          coef    std err          t      P>|t|      [0.025      0.975]\n",
            "---------------------------------------------------------------------------------------\n",
            "Intercept              97.3200      2.071     46.999      0.000      93.211     101.429\n",
            "condition[T.threat]   -11.0000      2.928     -3.756      0.000     -16.811      -5.189\n",
            "==============================================================================\n",
            "Omnibus:                        1.433   Durbin-Watson:                   2.052\n",
            "Prob(Omnibus):                  0.488   Jarque-Bera (JB):                1.011\n",
            "Skew:                           0.232   Prob(JB):                        0.603\n",
            "Kurtosis:                       3.164   Cond. No.                         2.62\n",
            "==============================================================================\n",
            "\n",
            "Notes:\n",
            "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
          ]
        }
      ],
      "source": [
        "# Mod_01 (Y~X) - IQ / Condição\n",
        "Mod_1 = smf.ols('IQ ~ condition', data=Threat).fit()\n",
        "print(Mod_1.params)\n",
        "print(Mod_1.summary())"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "_-IBukCwGVq2",
        "outputId": "11439ab1-26b8-499a-abd1-3e79a0aa5e80"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Intercept              55.997696\n",
            "condition[T.threat]    -2.407489\n",
            "WM                      0.752409\n",
            "dtype: float64\n",
            "                            OLS Regression Results                            \n",
            "==============================================================================\n",
            "Dep. Variable:                     IQ   R-squared:                       0.543\n",
            "Model:                            OLS   Adj. R-squared:                  0.533\n",
            "Method:                 Least Squares   F-statistic:                     57.59\n",
            "Date:                Sat, 29 Oct 2022   Prob (F-statistic):           3.27e-17\n",
            "Time:                        00:24:16   Log-Likelihood:                -376.86\n",
            "No. Observations:                 100   AIC:                             759.7\n",
            "Df Residuals:                      97   BIC:                             767.5\n",
            "Df Model:                           2                                         \n",
            "Covariance Type:            nonrobust                                         \n",
            "=======================================================================================\n",
            "                          coef    std err          t      P>|t|      [0.025      0.975]\n",
            "---------------------------------------------------------------------------------------\n",
            "Intercept              55.9977      4.644     12.058      0.000      46.781      65.215\n",
            "condition[T.threat]    -2.4075      2.316     -1.039      0.301      -7.005       2.190\n",
            "WM                      0.7524      0.080      9.406      0.000       0.594       0.911\n",
            "==============================================================================\n",
            "Omnibus:                        0.065   Durbin-Watson:                   2.189\n",
            "Prob(Omnibus):                  0.968   Jarque-Bera (JB):                0.004\n",
            "Skew:                          -0.002   Prob(JB):                        0.998\n",
            "Kurtosis:                       2.971   Cond. No.                         235.\n",
            "==============================================================================\n",
            "\n",
            "Notes:\n",
            "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
          ]
        }
      ],
      "source": [
        "# Mod_02 (Y~X+M) - IQ / Condição + WM\n",
        "Mod_2 = smf.ols('IQ ~ condition + WM', data=Threat).fit()\n",
        "print(Mod_2.params)\n",
        "print(Mod_2.summary())"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "PNHPIr0uGVmk",
        "outputId": "c9cd65f7-a309-481d-99f4-e2ced260f128"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Intercept              54.92\n",
            "condition[T.threat]   -11.42\n",
            "dtype: float64\n",
            "                            OLS Regression Results                            \n",
            "==============================================================================\n",
            "Dep. Variable:                     WM   R-squared:                       0.156\n",
            "Model:                            OLS   Adj. R-squared:                  0.147\n",
            "Method:                 Least Squares   F-statistic:                     18.05\n",
            "Date:                Sat, 29 Oct 2022   Prob (F-statistic):           4.91e-05\n",
            "Time:                        00:24:47   Log-Likelihood:                -400.71\n",
            "No. Observations:                 100   AIC:                             805.4\n",
            "Df Residuals:                      98   BIC:                             810.6\n",
            "Df Model:                           1                                         \n",
            "Covariance Type:            nonrobust                                         \n",
            "=======================================================================================\n",
            "                          coef    std err          t      P>|t|      [0.025      0.975]\n",
            "---------------------------------------------------------------------------------------\n",
            "Intercept              54.9200      1.901     28.895      0.000      51.148      58.692\n",
            "condition[T.threat]   -11.4200      2.688     -4.249      0.000     -16.754      -6.086\n",
            "==============================================================================\n",
            "Omnibus:                        1.342   Durbin-Watson:                   2.252\n",
            "Prob(Omnibus):                  0.511   Jarque-Bera (JB):                1.180\n",
            "Skew:                          -0.086   Prob(JB):                        0.554\n",
            "Kurtosis:                       2.497   Cond. No.                         2.62\n",
            "==============================================================================\n",
            "\n",
            "Notes:\n",
            "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
          ]
        }
      ],
      "source": [
        "# Mod_03 (M~X) - WM / condição\n",
        "Mod_3 = smf.ols('WM ~ condition', data=Threat).fit()\n",
        "print(Mod_3.params)\n",
        "print(Mod_3.summary())"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 581
        },
        "id": "go-TPjWZPF0M",
        "outputId": "45724bbd-0e3e-44f3-c455-437438e8ab5e"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 654.1x576 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Gráfico de Dispersão por condição\n",
        "graph2 = sns.lmplot( data=Threat, x='WM', y='IQ', hue='condition', height=8)\n",
        "# Nomeação dos eixos\n",
        "graph2.set_axis_labels('Working Memory Score', 'IQ Score')\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4ARJ1wWqGm4r"
      },
      "source": [
        "\n",
        "Compare as estatísticas de sumário de cada modelo e responda: há mediação da memória de trabalho?"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "M7cbizCrSGG4"
      },
      "source": [
        "Sim, Há mediação TOTAL de memória de trabalho. Isso porque, após a inserção da variável WM no modelo, o efeito da ameaça deixou de ser significativo. Além disso, um modelo de predição de 'WM' à partir da condição de ameaça pode ser feito. "
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "H6ERNX-gZWgp"
      },
      "source": [
        "# Referências:"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "q0SebLqnZcQA"
      },
      "source": [
        "[The General Linear Model](https://statsthinking21.github.io/statsthinking21-python/13-GeneralLinearModel.html)\n",
        "\n",
        "[Generalized Linear Models - Statsmodels Documentation](https://www.statsmodels.org/stable/glm.html)\n",
        "\n",
        "[6 Ways to Encode Features for Machine Learning Algorithms](https://towardsdatascience.com/6-ways-to-encode-features-for-machine-learning-algorithms-21593f6238b0)\n",
        "\n",
        "[O que é Modelagem de Equações Estruturais?](https://statplace.com.br/blog/modelagem-de-equacoes-estruturais/)\n"
      ]
    }
  ],
  "metadata": {
    "colab": {
      "collapsed_sections": [],
      "provenance": [],
      "toc_visible": true
    },
    "kernelspec": {
      "display_name": "Python 3.10.6 ('venv': venv)",
      "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": "3.10.6"
    },
    "orig_nbformat": 4,
    "vscode": {
      "interpreter": {
        "hash": "17a21100134af3592d09988eee871935e79466d45670c914d929fa5f969b25f9"
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}