LRMEval.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Untitled1.ipynb",
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "z65q4Ov7aPws"
      },
      "source": [
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "from sklearn.linear_model import LinearRegression\n",
        "from sklearn import datasets, linear_model\n",
        "from sklearn.metrics import *\n",
        "import statistics\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "from numpy.random import randn\n",
        "from numpy.random import seed\n",
        "from scipy.stats import pearsonr\n",
        "import numpy as np\n",
        "import statsmodels.api as sm\n",
        "from scipy.stats import gaussian_kde\n",
        "from numpy import linspace"
      ],
      "execution_count": 42,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VfhKMqLmbPb8"
      },
      "source": [
        "df = pd.read_csv(\"/content/drive/MyDrive/Colab Notebooks/df.csv\")"
      ],
      "execution_count": 43,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "KVHRtjTmbYTU",
        "outputId": "b1b2ff45-c973-4d68-f958-a2bbe1b9c976"
      },
      "source": [
        "print(df.head(10))\n",
        "\n",
        "sns.regplot(x=df[\"BMXWT\"], y=df[\"BMXHIP\"], line_kws={\"color\":\"r\",\"alpha\":0.7,\"lw\":5})\n",
        "plt.show()\n",
        "\n",
        "sns.regplot(x=df[\"BMXHT\"], y=df[\"BMXHIP\"], line_kws={\"color\":\"r\",\"alpha\":0.7,\"lw\":5})\n",
        "plt.show()\n",
        "\n",
        "sns.regplot(x=df[\"BMXWAIST\"], y=df[\"BMXHIP\"], line_kws={\"color\":\"r\",\"alpha\":0.7,\"lw\":5})\n",
        "plt.show()\n",
        "\n",
        "sns.regplot(x=df[\"BMXWT\"], y=df[\"BMXHIP\"], fit_reg=False)\n",
        "sns.regplot(x=df[\"BMXHT\"], y=df[\"BMXHIP\"], fit_reg=False)\n",
        "sns.regplot(x=df[\"BMXWAIST\"], y=df[\"BMXHIP\"], fit_reg=False)"
      ],
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "   BMXWT  BMXHT  BMXWAIST  BMXHIP\n",
            "0   57.9  168.2      68.0    66.6\n",
            "1   34.0  145.4      60.6    67.4\n",
            "2   52.5  169.3      68.0    68.7\n",
            "3   31.6  138.5      62.9    69.1\n",
            "4   30.7  140.5      59.2    69.7\n",
            "5   29.4  141.3      55.2    69.9\n",
            "6   36.5  150.5      59.8    69.9\n",
            "7   31.4  134.0      59.4    70.0\n",
            "8   29.2  137.5      56.4    70.1\n",
            "9   35.2  144.8      64.7    70.5\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f0ef77f7890>"
            ]
          },
          "metadata": {},
          "execution_count": 44
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "D4UIu4l1cBAn",
        "outputId": "25ff25eb-6ba6-4018-e2f8-cc061a5533b9"
      },
      "source": [
        "#CC between Hip circumference and weight:\n",
        "corr, _ = pearsonr(df[\"BMXHIP\"], df[\"BMXWT\"])\n",
        "print('Pearsons correlation: %.3f' % corr)\n",
        "\n",
        "#CC between Hip circumference and height:\n",
        "corr, _ = pearsonr(df[\"BMXHIP\"], df[\"BMXHT\"])\n",
        "print('Pearsons correlation: %.3f' % corr)\n",
        "\n",
        "#CC between Hip circumference and Waist circumference:\n",
        "corr, _ = pearsonr(df[\"BMXHIP\"], df[\"BMXWAIST\"])\n",
        "print('Pearsons correlation: %.3f' % corr)"
      ],
      "execution_count": 45,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Pearsons correlation: 0.879\n",
            "Pearsons correlation: 0.108\n",
            "Pearsons correlation: 0.876\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "AXuX-1MXcrqA"
      },
      "source": [
        "#Regression Models:\n",
        "y = df['BMXHIP']"
      ],
      "execution_count": 46,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nv8TfaqTccQ-",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "606a4ddb-dc64-42f8-e6f9-62e5fe0a474a"
      },
      "source": [
        "#Model 1\n",
        "X1 = df[['BMXHT']]\n",
        "\n",
        "model_1 = linear_model.LinearRegression()\n",
        "model_1.fit(X1, y)\n",
        "\n",
        "print(model_1.coef_)\n",
        "print(model_1.intercept_)\n",
        "\n",
        "r_sq_model_1 = model_1.score(X1, y)\n",
        "print(r_sq_model_1)\n",
        "\n",
        "y1_pred = model_1.predict(X1)\n",
        "print('predicted response:', y1_pred, sep='\\n')\n",
        "\n",
        "X1a = sm.add_constant(X1)\n",
        "model_1a = sm.OLS(y, X1a)\n",
        "\n",
        "results_model_1a = model_1a.fit()\n",
        "print(results_model_1a.summary())\n",
        "\n",
        "#Residual Standard Error (RSE):\n",
        "model_1a = sm.OLS(y, X1a).fit()\n",
        "model_1a.resid.std(ddof=X1a.shape[1])\n",
        "\n",
        "#Bland-Altman Plot:\n",
        "df['BMXHIP_1'] = 78.90301 + (0.16066*df['BMXHT'])\n",
        "df['md_1'] = df['BMXHIP']-df['BMXHIP_1']\n",
        "print(df.head(10))\n",
        "\n",
        "df['mean_BMXHIP'] = statistics.mean(df['BMXHIP'])\n",
        "df['sd_BMXHIP'] = statistics.stdev(df['BMXHIP'])\n",
        "df['mean_BMXHIP_1'] = statistics.mean(df['BMXHIP_1'])\n",
        "df['sd_BMXHIP_1'] = statistics.stdev(df['BMXHIP_1'])\n",
        "df['mean_md_1'] = statistics.mean(df['md_1'])\n",
        "df['sd_md_1'] = statistics.stdev(df['md_1'])\n",
        "\n",
        "#Plot drawn:\n",
        "print(\"Mean md_1:\", statistics.mean(df['md_1']))\n",
        "print(\"SD md_1: \", statistics.stdev(df['md_1']))\n",
        "print(\"LLA: \", statistics.mean(df['md_1'])-(1.96*statistics.stdev(df['md_1'])))\n",
        "print(\"ULA: \", statistics.mean(df['md_1'])+(1.96*statistics.stdev(df['md_1'])))\n",
        "\n",
        "plt.plot( 'BMXHIP', 'md_1', data=df, linestyle='none', marker='o')\n",
        "plt.axhline(y = 0, color = 'black', linestyle = '-')\n",
        "plt.axhline(y = -0.00043955585024463647, color = 'blue', linestyle = '--')\n",
        "plt.axhline(y = -29.126151124880025, color = 'red', linestyle = '--')\n",
        "plt.axhline(y = 29.125272013179533, color = 'red', linestyle = '--')\n",
        "plt.show()\n",
        "\n",
        "#Limits of Agreement amplitude:\n",
        "LOAA_1 = 29.126151124880025 + 29.125272013179533\n",
        "print(\"LOAA_1:\" , LOAA_1)"
      ],
      "execution_count": 47,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[0.16065735]\n",
            "78.90301226720597\n",
            "0.011748382687132297\n",
            "predicted response:\n",
            "[105.9255779  102.2625904  106.10230098 ... 105.89344643 104.65638486\n",
            " 104.59212192]\n",
            "                            OLS Regression Results                            \n",
            "==============================================================================\n",
            "Dep. Variable:                 BMXHIP   R-squared:                       0.012\n",
            "Model:                            OLS   Adj. R-squared:                  0.012\n",
            "Method:                 Least Squares   F-statistic:                     116.5\n",
            "Date:                Mon, 22 Nov 2021   Prob (F-statistic):           5.20e-27\n",
            "Time:                        21:12:41   Log-Likelihood:                -40364.\n",
            "No. Observations:                9803   AIC:                         8.073e+04\n",
            "Df Residuals:                    9801   BIC:                         8.075e+04\n",
            "Df Model:                           1                                         \n",
            "Covariance Type:            nonrobust                                         \n",
            "==============================================================================\n",
            "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const         78.9030      2.482     31.784      0.000      74.037      83.769\n",
            "BMXHT          0.1607      0.015     10.794      0.000       0.131       0.190\n",
            "==============================================================================\n",
            "Omnibus:                     1661.065   Durbin-Watson:                   0.022\n",
            "Prob(Omnibus):                  0.000   Jarque-Bera (JB):             3078.177\n",
            "Skew:                           1.064   Prob(JB):                         0.00\n",
            "Kurtosis:                       4.735   Cond. No.                     2.76e+03\n",
            "==============================================================================\n",
            "\n",
            "Warnings:\n",
            "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
            "[2] The condition number is large, 2.76e+03. This might indicate that there are\n",
            "strong multicollinearity or other numerical problems.\n",
            "   BMXWT  BMXHT  BMXWAIST  BMXHIP    BMXHIP_1       md_1\n",
            "0   57.9  168.2      68.0    66.6  105.926022 -39.326022\n",
            "1   34.0  145.4      60.6    67.4  102.262974 -34.862974\n",
            "2   52.5  169.3      68.0    68.7  106.102748 -37.402748\n",
            "3   31.6  138.5      62.9    69.1  101.154420 -32.054420\n",
            "4   30.7  140.5      59.2    69.7  101.475740 -31.775740\n",
            "5   29.4  141.3      55.2    69.9  101.604268 -31.704268\n",
            "6   36.5  150.5      59.8    69.9  103.082340 -33.182340\n",
            "7   31.4  134.0      59.4    70.0  100.431450 -30.431450\n",
            "8   29.2  137.5      56.4    70.1  100.993760 -30.893760\n",
            "9   35.2  144.8      64.7    70.5  102.166578 -31.666578\n",
            "Mean md_1: -0.00043955585024463647\n",
            "SD md_1:  14.860056922974378\n",
            "LLA:  -29.126151124880025\n",
            "ULA:  29.125272013179533\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "LOAA_1: 58.25142313805956\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "QV1Ofsjgdqn4",
        "outputId": "65acb3d1-79fd-409e-9489-4bb14c54569d"
      },
      "source": [
        "#Model 2\n",
        "X2 = df[['BMXWT']]\n",
        "\n",
        "model_2 = linear_model.LinearRegression()\n",
        "model_2.fit(X2, y)\n",
        "\n",
        "print(model_2.coef_)\n",
        "print(model_2.intercept_)\n",
        "\n",
        "#Add a constant:\n",
        "X2a = sm.add_constant(X2)\n",
        "model_2a = sm.OLS(y, X2a)\n",
        "\n",
        "results_model_2a = model_2a.fit()\n",
        "print(results_model_2a.summary())\n",
        "\n",
        "#Residual Standard Error (RSE):\n",
        "model_2a = sm.OLS(y, X2a).fit()\n",
        "model_2a.resid.std(ddof=X2a.shape[1])\n",
        "\n",
        "#Bland-Altman Plot:\n",
        "df['BMXHIP_2'] = 59.709755 + (0.568801*df['BMXWT'])\n",
        "df['md_2'] = df['BMXHIP']-df['BMXHIP_2']\n",
        "print(df.head(10))\n",
        "\n",
        "df['mean_BMXHIP_2'] = statistics.mean(df['BMXHIP_2'])\n",
        "df['sd_BMXHIP_2'] = statistics.stdev(df['BMXHIP_2'])\n",
        "df['mean_md_2'] = statistics.mean(df['md_2'])\n",
        "df['sd_md_2'] = statistics.stdev(df['md_2'])\n",
        "print(df.head(10))\n",
        "\n",
        "#Plot drawn:\n",
        "print(\"Mean md_2:\", statistics.mean(df['md_2']))\n",
        "print(\"SD md_2: \", statistics.stdev(df['md_2']))\n",
        "print(\"LLA: \", statistics.mean(df['md_2'])-(1.96*statistics.stdev(df['md_2'])))\n",
        "print(\"ULA: \", statistics.mean(df['md_2'])+(1.96*statistics.stdev(df['md_2'])))\n",
        "\n",
        "plt.plot( 'BMXHIP', 'md_2', data=df, linestyle='none', marker='o')\n",
        "plt.axhline(y = 0, color = 'black', linestyle = '-')\n",
        "plt.axhline(y = -6.812200491674594e-08, color = 'blue', linestyle = '-')\n",
        "plt.axhline(y = -13.974429503341867, color = 'red', linestyle = '-')\n",
        "plt.axhline(y = 13.974429367097859, color = 'red', linestyle = '-')\n",
        "plt.show()\n",
        "\n",
        "#Limits of Agreement amplitude:\n",
        "LOAA_2 = 13.974429503341867 + 13.974429367097859\n",
        "print(\"LOAA_2:\" , LOAA_2)"
      ],
      "execution_count": 48,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[0.568801]\n",
            "59.70975520832089\n",
            "                            OLS Regression Results                            \n",
            "==============================================================================\n",
            "Dep. Variable:                 BMXHIP   R-squared:                       0.772\n",
            "Model:                            OLS   Adj. R-squared:                  0.772\n",
            "Method:                 Least Squares   F-statistic:                 3.328e+04\n",
            "Date:                Mon, 22 Nov 2021   Prob (F-statistic):               0.00\n",
            "Time:                        21:12:42   Log-Likelihood:                -33165.\n",
            "No. Observations:                9803   AIC:                         6.633e+04\n",
            "Df Residuals:                    9801   BIC:                         6.635e+04\n",
            "Df Model:                           1                                         \n",
            "Covariance Type:            nonrobust                                         \n",
            "==============================================================================\n",
            "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const         59.7098      0.262    227.967      0.000      59.196      60.223\n",
            "BMXWT          0.5688      0.003    182.429      0.000       0.563       0.575\n",
            "==============================================================================\n",
            "Omnibus:                      536.287   Durbin-Watson:                   1.223\n",
            "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              647.719\n",
            "Skew:                           0.568   Prob(JB):                    2.24e-141\n",
            "Kurtosis:                       3.545   Cond. No.                         306.\n",
            "==============================================================================\n",
            "\n",
            "Warnings:\n",
            "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
            "   BMXWT  BMXHT  BMXWAIST  BMXHIP  ...  mean_md_1    sd_md_1   BMXHIP_2       md_2\n",
            "0   57.9  168.2      68.0    66.6  ...   -0.00044  14.860057  92.643333 -26.043333\n",
            "1   34.0  145.4      60.6    67.4  ...   -0.00044  14.860057  79.048989 -11.648989\n",
            "2   52.5  169.3      68.0    68.7  ...   -0.00044  14.860057  89.571808 -20.871808\n",
            "3   31.6  138.5      62.9    69.1  ...   -0.00044  14.860057  77.683867  -8.583867\n",
            "4   30.7  140.5      59.2    69.7  ...   -0.00044  14.860057  77.171946  -7.471946\n",
            "5   29.4  141.3      55.2    69.9  ...   -0.00044  14.860057  76.432504  -6.532504\n",
            "6   36.5  150.5      59.8    69.9  ...   -0.00044  14.860057  80.470991 -10.570991\n",
            "7   31.4  134.0      59.4    70.0  ...   -0.00044  14.860057  77.570106  -7.570106\n",
            "8   29.2  137.5      56.4    70.1  ...   -0.00044  14.860057  76.318744  -6.218744\n",
            "9   35.2  144.8      64.7    70.5  ...   -0.00044  14.860057  79.731550  -9.231550\n",
            "\n",
            "[10 rows x 14 columns]\n",
            "   BMXWT  BMXHT  BMXWAIST  ...  sd_BMXHIP_2     mean_md_2   sd_md_2\n",
            "0   57.9  168.2      68.0  ...    13.138197 -6.812200e-08  7.129811\n",
            "1   34.0  145.4      60.6  ...    13.138197 -6.812200e-08  7.129811\n",
            "2   52.5  169.3      68.0  ...    13.138197 -6.812200e-08  7.129811\n",
            "3   31.6  138.5      62.9  ...    13.138197 -6.812200e-08  7.129811\n",
            "4   30.7  140.5      59.2  ...    13.138197 -6.812200e-08  7.129811\n",
            "5   29.4  141.3      55.2  ...    13.138197 -6.812200e-08  7.129811\n",
            "6   36.5  150.5      59.8  ...    13.138197 -6.812200e-08  7.129811\n",
            "7   31.4  134.0      59.4  ...    13.138197 -6.812200e-08  7.129811\n",
            "8   29.2  137.5      56.4  ...    13.138197 -6.812200e-08  7.129811\n",
            "9   35.2  144.8      64.7  ...    13.138197 -6.812200e-08  7.129811\n",
            "\n",
            "[10 rows x 18 columns]\n",
            "Mean md_2: -6.812200491674594e-08\n",
            "SD md_2:  7.129810936336665\n",
            "LLA:  -13.974429503341867\n",
            "ULA:  13.974429367097859\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "LOAA_2: 27.948858870439725\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "4pnizyo2d3il",
        "outputId": "5f604b95-7b50-4506-9e65-579372d45a16"
      },
      "source": [
        "#Model 3\n",
        "X3 = df[['BMXWAIST']]\n",
        "\n",
        "model_3 = linear_model.LinearRegression()\n",
        "model_3.fit(X3, y)\n",
        "\n",
        "print(model_3.coef_)\n",
        "print(model_3.intercept_)\n",
        "\n",
        "#Add a constant:\n",
        "X3a = sm.add_constant(X3)\n",
        "model_3a = sm.OLS(y, X3a)\n",
        "\n",
        "results_model_3a = model_3a.fit()\n",
        "print(results_model_3a.summary())\n",
        "\n",
        "#Residual Standard Error (RSE):\n",
        "model_3a = sm.OLS(y, X3a).fit()\n",
        "model_3a.resid.std(ddof=X3a.shape[1])\n",
        "\n",
        "#Bland-Altman Plot:\n",
        "df['BMXHIP_3'] = 35.59334 + (0.71649*df['BMXWAIST'])\n",
        "df['md_3'] = df['BMXHIP']-df['BMXHIP_3']\n",
        "print(df.head(10))\n",
        "\n",
        "df['mean_BMXHIP_3'] = statistics.mean(df['BMXHIP_3'])\n",
        "df['sd_BMXHIP_3'] = statistics.stdev(df['BMXHIP_3'])\n",
        "df['mean_md_3'] = statistics.mean(df['md_3'])\n",
        "df['sd_md_3'] = statistics.stdev(df['md_3'])\n",
        "print(df.head(10))\n",
        "\n",
        "#Plot drawn:\n",
        "print(\"Mean md_3:\", statistics.mean(df['md_3']))\n",
        "print(\"SD md_3: \", statistics.stdev(df['md_3']))\n",
        "print(\"LLA: \", statistics.mean(df['md_3'])-(1.96*statistics.stdev(df['md_3'])))\n",
        "print(\"ULA: \", statistics.mean(df['md_3'])+(1.96*statistics.stdev(df['md_3'])))\n",
        "\n",
        "plt.plot( 'BMXHIP', 'md_3', data=df, linestyle='none', marker='o')\n",
        "plt.axhline(y = 0, color = 'black', linestyle = '-')\n",
        "plt.axhline(y = 0.00031253381618476223, color = 'blue', linestyle = '--')\n",
        "plt.axhline(y = -14.118955041009063, color = 'red', linestyle = '--')\n",
        "plt.axhline(y = 14.119580108641433, color = 'red', linestyle = '--')\n",
        "plt.show()\n",
        "\n",
        "#Limits of Agreement amplitude:\n",
        "LOAA_3 = 14.118955041009063 + 14.119580108641433\n",
        "print(\"LOAA_3:\" , LOAA_3)"
      ],
      "execution_count": 49,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[0.7164932]\n",
            "35.593339970401516\n",
            "                            OLS Regression Results                            \n",
            "==============================================================================\n",
            "Dep. Variable:                 BMXHIP   R-squared:                       0.768\n",
            "Model:                            OLS   Adj. R-squared:                  0.768\n",
            "Method:                 Least Squares   F-statistic:                 3.240e+04\n",
            "Date:                Mon, 22 Nov 2021   Prob (F-statistic):               0.00\n",
            "Time:                        21:12:43   Log-Likelihood:                -33266.\n",
            "No. Observations:                9803   AIC:                         6.654e+04\n",
            "Df Residuals:                    9801   BIC:                         6.655e+04\n",
            "Df Model:                           1                                         \n",
            "Covariance Type:            nonrobust                                         \n",
            "==============================================================================\n",
            "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const         35.5933      0.396     89.895      0.000      34.817      36.369\n",
            "BMXWAIST       0.7165      0.004    180.002      0.000       0.709       0.724\n",
            "==============================================================================\n",
            "Omnibus:                      873.176   Durbin-Watson:                   0.619\n",
            "Prob(Omnibus):                  0.000   Jarque-Bera (JB):             1481.122\n",
            "Skew:                           0.646   Prob(JB):                         0.00\n",
            "Kurtosis:                       4.398   Cond. No.                         541.\n",
            "==============================================================================\n",
            "\n",
            "Warnings:\n",
            "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
            "   BMXWT  BMXHT  BMXWAIST  BMXHIP  ...     mean_md_2   sd_md_2   BMXHIP_3       md_3\n",
            "0   57.9  168.2      68.0    66.6  ... -6.812200e-08  7.129811  84.314660 -17.714660\n",
            "1   34.0  145.4      60.6    67.4  ... -6.812200e-08  7.129811  79.012634 -11.612634\n",
            "2   52.5  169.3      68.0    68.7  ... -6.812200e-08  7.129811  84.314660 -15.614660\n",
            "3   31.6  138.5      62.9    69.1  ... -6.812200e-08  7.129811  80.660561 -11.560561\n",
            "4   30.7  140.5      59.2    69.7  ... -6.812200e-08  7.129811  78.009548  -8.309548\n",
            "5   29.4  141.3      55.2    69.9  ... -6.812200e-08  7.129811  75.143588  -5.243588\n",
            "6   36.5  150.5      59.8    69.9  ... -6.812200e-08  7.129811  78.439442  -8.539442\n",
            "7   31.4  134.0      59.4    70.0  ... -6.812200e-08  7.129811  78.152846  -8.152846\n",
            "8   29.2  137.5      56.4    70.1  ... -6.812200e-08  7.129811  76.003376  -5.903376\n",
            "9   35.2  144.8      64.7    70.5  ... -6.812200e-08  7.129811  81.950243 -11.450243\n",
            "\n",
            "[10 rows x 20 columns]\n",
            "   BMXWT  BMXHT  BMXWAIST  ...  sd_BMXHIP_3  mean_md_3   sd_md_3\n",
            "0   57.9  168.2      68.0  ...    13.097766   0.000313  7.203708\n",
            "1   34.0  145.4      60.6  ...    13.097766   0.000313  7.203708\n",
            "2   52.5  169.3      68.0  ...    13.097766   0.000313  7.203708\n",
            "3   31.6  138.5      62.9  ...    13.097766   0.000313  7.203708\n",
            "4   30.7  140.5      59.2  ...    13.097766   0.000313  7.203708\n",
            "5   29.4  141.3      55.2  ...    13.097766   0.000313  7.203708\n",
            "6   36.5  150.5      59.8  ...    13.097766   0.000313  7.203708\n",
            "7   31.4  134.0      59.4  ...    13.097766   0.000313  7.203708\n",
            "8   29.2  137.5      56.4  ...    13.097766   0.000313  7.203708\n",
            "9   35.2  144.8      64.7  ...    13.097766   0.000313  7.203708\n",
            "\n",
            "[10 rows x 24 columns]\n",
            "Mean md_3: 0.00031253381618476223\n",
            "SD md_3:  7.203707946339412\n",
            "LLA:  -14.118955041009063\n",
            "ULA:  14.119580108641433\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "LOAA_3: 28.238535149650495\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "74reX-oHeIuL",
        "outputId": "afabc521-0929-4e8d-c05b-fd88fcf90547"
      },
      "source": [
        "#Model 4\n",
        "X4 = df[['BMXWT', 'BMXHT']]\n",
        "\n",
        "model_4 = linear_model.LinearRegression()\n",
        "model_4.fit(X4, y)\n",
        "\n",
        "print(model_4.coef_)\n",
        "print(model_4.intercept_)\n",
        "\n",
        "#Add a constant:\n",
        "X4a = sm.add_constant(X4)\n",
        "model_4a = sm.OLS(y, X4a)\n",
        "\n",
        "results_model_4a = model_4a.fit()\n",
        "print(results_model_4a.summary())\n",
        "\n",
        "#Residual Standard Error (RSE):\n",
        "model_4a = sm.OLS(y, X4a).fit()\n",
        "model_4a.resid.std(ddof=X4a.shape[1])\n",
        "\n",
        "#Bland-Altman Plot:\n",
        "df['BMXHIP_4'] = 137.920164 + (-0.518439*df['BMXHT']) + (0.669106*df['BMXWT'])\n",
        "df['md_4'] = df['BMXHIP']-df['BMXHIP_4']\n",
        "print(df.head(10))\n",
        "\n",
        "df['mean_BMXHIP_4'] = statistics.mean(df['BMXHIP_4'])\n",
        "df['sd_BMXHIP_4'] = statistics.stdev(df['BMXHIP_4'])\n",
        "df['mean_md_4'] = statistics.mean(df['md_4'])\n",
        "df['sd_md_4'] = statistics.stdev(df['md_4'])\n",
        "print(df.head(10))\n",
        "\n",
        "#Plot drawn:\n",
        "print(\"Mean md_4:\", statistics.mean(df['md_4']))\n",
        "print(\"SD md_4: \", statistics.stdev(df['md_4']))\n",
        "print(\"LLA: \", statistics.mean(df['md_4'])-(1.96*statistics.stdev(df['md_4'])))\n",
        "print(\"ULA: \", statistics.mean(df['md_4'])+(1.96*statistics.stdev(df['md_4'])))\n",
        "\n",
        "plt.plot( 'BMXHIP', 'md_4', data=df, linestyle='none', marker='o')\n",
        "plt.axhline(y = 0, color = 'black', linestyle = '-')\n",
        "plt.axhline(y = 0.0014407432010600173, color = 'blue', linestyle = '--')\n",
        "plt.axhline(y = -10.529108193463623, color = 'red', linestyle = '--')\n",
        "plt.axhline(y = 10.531989679865744, color = 'red', linestyle = '--')\n",
        "plt.show()\n",
        "\n",
        "#Limits of Agreement amplitude:\n",
        "LOAA_4 = 10.529108193463623 + 10.531989679865744\n",
        "print(\"LOAA_4:\" , LOAA_4)"
      ],
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[ 0.66910558 -0.51843014]\n",
            "137.92016401571618\n",
            "                            OLS Regression Results                            \n",
            "==============================================================================\n",
            "Dep. Variable:                 BMXHIP   R-squared:                       0.871\n",
            "Model:                            OLS   Adj. R-squared:                  0.871\n",
            "Method:                 Least Squares   F-statistic:                 3.303e+04\n",
            "Date:                Mon, 22 Nov 2021   Prob (F-statistic):               0.00\n",
            "Time:                        21:12:43   Log-Likelihood:                -30391.\n",
            "No. Observations:                9803   AIC:                         6.079e+04\n",
            "Df Residuals:                    9800   BIC:                         6.081e+04\n",
            "Df Model:                           2                                         \n",
            "Covariance Type:            nonrobust                                         \n",
            "==============================================================================\n",
            "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const        137.9202      0.927    148.799      0.000     136.103     139.737\n",
            "BMXWT          0.6691      0.003    255.281      0.000       0.664       0.674\n",
            "BMXHT         -0.5184      0.006    -86.360      0.000      -0.530      -0.507\n",
            "==============================================================================\n",
            "Omnibus:                      178.006   Durbin-Watson:                   1.283\n",
            "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              286.830\n",
            "Skew:                           0.174   Prob(JB):                     5.20e-63\n",
            "Kurtosis:                       3.762   Cond. No.                     3.17e+03\n",
            "==============================================================================\n",
            "\n",
            "Warnings:\n",
            "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
            "[2] The condition number is large, 3.17e+03. This might indicate that there are\n",
            "strong multicollinearity or other numerical problems.\n",
            "   BMXWT  BMXHT  BMXWAIST  BMXHIP  ...  mean_md_3   sd_md_3   BMXHIP_4       md_4\n",
            "0   57.9  168.2      68.0    66.6  ...   0.000313  7.203708  89.459962 -22.859962\n",
            "1   34.0  145.4      60.6    67.4  ...   0.000313  7.203708  85.288737 -17.888737\n",
            "2   52.5  169.3      68.0    68.7  ...   0.000313  7.203708  85.276506 -16.576506\n",
            "3   31.6  138.5      62.9    69.1  ...   0.000313  7.203708  87.260112 -18.160112\n",
            "4   30.7  140.5      59.2    69.7  ...   0.000313  7.203708  85.621039 -15.921039\n",
            "5   29.4  141.3      55.2    69.9  ...   0.000313  7.203708  84.336450 -14.436450\n",
            "6   36.5  150.5      59.8    69.9  ...   0.000313  7.203708  84.317464 -14.417463\n",
            "7   31.4  134.0      59.4    70.0  ...   0.000313  7.203708  89.459266 -19.459266\n",
            "8   29.2  137.5      56.4    70.1  ...   0.000313  7.203708  86.172697 -16.072697\n",
            "9   35.2  144.8      64.7    70.5  ...   0.000313  7.203708  86.402728 -15.902728\n",
            "\n",
            "[10 rows x 26 columns]\n",
            "   BMXWT  BMXHT  BMXWAIST  ...  sd_BMXHIP_4  mean_md_4   sd_md_4\n",
            "0   57.9  168.2      68.0  ...      13.9492   0.001441  5.372729\n",
            "1   34.0  145.4      60.6  ...      13.9492   0.001441  5.372729\n",
            "2   52.5  169.3      68.0  ...      13.9492   0.001441  5.372729\n",
            "3   31.6  138.5      62.9  ...      13.9492   0.001441  5.372729\n",
            "4   30.7  140.5      59.2  ...      13.9492   0.001441  5.372729\n",
            "5   29.4  141.3      55.2  ...      13.9492   0.001441  5.372729\n",
            "6   36.5  150.5      59.8  ...      13.9492   0.001441  5.372729\n",
            "7   31.4  134.0      59.4  ...      13.9492   0.001441  5.372729\n",
            "8   29.2  137.5      56.4  ...      13.9492   0.001441  5.372729\n",
            "9   35.2  144.8      64.7  ...      13.9492   0.001441  5.372729\n",
            "\n",
            "[10 rows x 30 columns]\n",
            "Mean md_4: 0.0014407432010600173\n",
            "SD md_4:  5.372729049318716\n",
            "LLA:  -10.529108193463623\n",
            "ULA:  10.531989679865744\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "LOAA_4: 21.061097873329366\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "AJL0cx5CeScM",
        "outputId": "b96cf412-0af8-4806-db37-c5030ed69d8a"
      },
      "source": [
        "#Model 5\n",
        "X5 = df[['BMXHT', 'BMXWAIST']]\n",
        "\n",
        "model_5 = linear_model.LinearRegression()\n",
        "model_5.fit(X5, y)\n",
        "\n",
        "print(model_5.coef_)\n",
        "print(model_5.intercept_)\n",
        "\n",
        "#Add a constant:\n",
        "X5a = sm.add_constant(X5)\n",
        "model_5a = sm.OLS(y, X5a)\n",
        "\n",
        "results_model_5a = model_5a.fit()\n",
        "print(results_model_5a.summary())\n",
        "\n",
        "#Residual Standard Error (RSE):\n",
        "model_5a = sm.OLS(y, X5a).fit()\n",
        "model_5a.resid.std(ddof=X5a.shape[1])\n",
        "\n",
        "#Bland-Altman Plot:\n",
        "df['BMXHIP_5'] = 50.561862 + (-0.095965*df['BMXHT']) + (0.726806*df['BMXWAIST'])\n",
        "df['md_5'] = df['BMXHIP']-df['BMXHIP_5']\n",
        "print(df.head(10))\n",
        "\n",
        "df['mean_BMXHIP_5'] = statistics.mean(df['BMXHIP_5'])\n",
        "df['sd_BMXHIP_5'] = statistics.stdev(df['BMXHIP_5'])\n",
        "df['mean_md_5'] = statistics.mean(df['md_5'])\n",
        "df['sd_md_5'] = statistics.stdev(df['md_5'])\n",
        "print(df.head(10))\n",
        "\n",
        "#Plot drawn:\n",
        "print(\"Mean md_5:\", statistics.mean(df['md_5']))\n",
        "print(\"SD md_5: \", statistics.stdev(df['md_5']))\n",
        "print(\"LLA: \", statistics.mean(df['md_5'])-(1.96*statistics.stdev(df['md_5'])))\n",
        "print(\"ULA: \", statistics.mean(df['md_5'])+(1.96*statistics.stdev(df['md_5'])))\n",
        "\n",
        "plt.plot( 'BMXHIP', 'md_5', data=df, linestyle='none', marker='o')\n",
        "plt.axhline(y = 0, color = 'black', linestyle = '-')\n",
        "plt.axhline(y = 1.7583862088800805e-05, color = 'blue', linestyle = '--')\n",
        "plt.axhline(y = -13.99612617351395, color = 'red', linestyle = '--')\n",
        "plt.axhline(y = 13.996161341238128, color = 'red', linestyle = '--')\n",
        "plt.show()\n",
        "\n",
        "#Limits of Agreement amplitude:\n",
        "LOAA_5 = 13.99612617351395 + 13.996161341238128\n",
        "print(\"LOAA_5:\" , LOAA_5)"
      ],
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[-0.09596471  0.72680568]\n",
            "50.56186191985632\n",
            "                            OLS Regression Results                            \n",
            "==============================================================================\n",
            "Dep. Variable:                 BMXHIP   R-squared:                       0.772\n",
            "Model:                            OLS   Adj. R-squared:                  0.772\n",
            "Method:                 Least Squares   F-statistic:                 1.657e+04\n",
            "Date:                Mon, 22 Nov 2021   Prob (F-statistic):               0.00\n",
            "Time:                        21:12:44   Log-Likelihood:                -33180.\n",
            "No. Observations:                9803   AIC:                         6.637e+04\n",
            "Df Residuals:                    9800   BIC:                         6.639e+04\n",
            "Df Model:                           2                                         \n",
            "Covariance Type:            nonrobust                                         \n",
            "==============================================================================\n",
            "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const         50.5619      1.203     42.021      0.000      48.203      52.921\n",
            "BMXHT         -0.0960      0.007    -13.160      0.000      -0.110      -0.082\n",
            "BMXWAIST       0.7268      0.004    180.662      0.000       0.719       0.735\n",
            "==============================================================================\n",
            "Omnibus:                      772.729   Durbin-Watson:                   0.639\n",
            "Prob(Omnibus):                  0.000   Jarque-Bera (JB):             1266.893\n",
            "Skew:                           0.598   Prob(JB):                    7.90e-276\n",
            "Kurtosis:                       4.293   Cond. No.                     3.23e+03\n",
            "==============================================================================\n",
            "\n",
            "Warnings:\n",
            "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
            "[2] The condition number is large, 3.23e+03. This might indicate that there are\n",
            "strong multicollinearity or other numerical problems.\n",
            "   BMXWT  BMXHT  BMXWAIST  BMXHIP  ...  mean_md_4   sd_md_4   BMXHIP_5       md_5\n",
            "0   57.9  168.2      68.0    66.6  ...   0.001441  5.372729  83.843357 -17.243357\n",
            "1   34.0  145.4      60.6    67.4  ...   0.001441  5.372729  80.652995 -13.252995\n",
            "2   52.5  169.3      68.0    68.7  ...   0.001441  5.372729  83.737796 -15.037796\n",
            "3   31.6  138.5      62.9    69.1  ...   0.001441  5.372729  82.986807 -13.886807\n",
            "4   30.7  140.5      59.2    69.7  ...   0.001441  5.372729  80.105695 -10.405695\n",
            "5   29.4  141.3      55.2    69.9  ...   0.001441  5.372729  77.121699  -7.221699\n",
            "6   36.5  150.5      59.8    69.9  ...   0.001441  5.372729  79.582128  -9.682128\n",
            "7   31.4  134.0      59.4    70.0  ...   0.001441  5.372729  80.874828 -10.874828\n",
            "8   29.2  137.5      56.4    70.1  ...   0.001441  5.372729  78.358533  -8.258533\n",
            "9   35.2  144.8      64.7    70.5  ...   0.001441  5.372729  83.690478 -13.190478\n",
            "\n",
            "[10 rows x 32 columns]\n",
            "   BMXWT  BMXHT  BMXWAIST  ...  sd_BMXHIP_5  mean_md_5  sd_md_5\n",
            "0   57.9  168.2      68.0  ...    13.132184   0.000018  7.14089\n",
            "1   34.0  145.4      60.6  ...    13.132184   0.000018  7.14089\n",
            "2   52.5  169.3      68.0  ...    13.132184   0.000018  7.14089\n",
            "3   31.6  138.5      62.9  ...    13.132184   0.000018  7.14089\n",
            "4   30.7  140.5      59.2  ...    13.132184   0.000018  7.14089\n",
            "5   29.4  141.3      55.2  ...    13.132184   0.000018  7.14089\n",
            "6   36.5  150.5      59.8  ...    13.132184   0.000018  7.14089\n",
            "7   31.4  134.0      59.4  ...    13.132184   0.000018  7.14089\n",
            "8   29.2  137.5      56.4  ...    13.132184   0.000018  7.14089\n",
            "9   35.2  144.8      64.7  ...    13.132184   0.000018  7.14089\n",
            "\n",
            "[10 rows x 36 columns]\n",
            "Mean md_5: 1.7583862088800805e-05\n",
            "SD md_5:  7.1408896721306325\n",
            "LLA:  -13.99612617351395\n",
            "ULA:  13.996161341238128\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXkAAAD4CAYAAAAJmJb0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO29fZwU1Zno/326p4EeVAaEGB1AUSMkXCIoVzH8NokvKyZKnCgRE81P7+YuuWaTXySGBIwJsOsLCXEld7Prxr3Zn250dVTMiGJCgpLkXhM0gzNIiJAYFaHVCMJodEbomTn3j+5qqquraqq7q6bfnu/nM9Bdp17O6a5+6jnPeV7EGIOiKIpSn8Qq3QFFURQlOlTIK4qi1DEq5BVFUeoYFfKKoih1jAp5RVGUOqap0h2wM378eHPCCSdUuhuKoig1xZYtW/YZYya4tVWVkD/hhBPo7OysdDcURVFqChHZ5dWm5hpFUZQ6RoW8oihKHaNCXlEUpY5RIa8oilLHqJBXFEWpY6rKu0ZRFKXR6OhKsXrDTl7p6eO4liRL5k2lbVZraOdXIa8oilIhOrpSLHtoG33pAQBSPX0se2gbQGiCXs01iqIoFWL1hp05AW/Rlx5g9YadoV0jNE1eROJAJ5AyxlwkIlOA+4CjgS3AZ40xh8K6nqIohUQ99VfC5ZWevqK2l0KYmvyXgeds778N3GaMORk4AHwuxGspiuLAmvqnevowHJ76d3SlKt01xYPjWpJFbS+FUIS8iEwELgT+V/a9AOcAD2Z3uQtoC+NaiqK4MxxTfyVclsybSjIRz9uWTMRZMm9qaNcIy1yzBvgacGT2/dFAjzGmP/t+D6BzRkWJkOGY+ivhYpnSqtq7RkQuAl43xmwRkY+WcPwiYBHA5MmTy+2OojQsx7UkSbkI9DCn/kr4tM1qjXTdJAxzzVzgEyLyEpmF1nOA7wEtImI9RCYCroZBY8wdxpjZxpjZEya4ZspUFCUAwzH1L5aOrhRzVz3BlKXrmbvqCV0fqABla/LGmGXAMoCsJv9VY8wVIvIAsICM4L8KeLjcaymK4s1wTP2L8d4ZDh9wZWiiDIb6OnCfiNwIdAE/jPBaiqIQ7dS/WKHttxBcC0K+XtxRQxXyxphfAL/Ivn4BOCPM8yuKEg6lCLBihXYtLwTX0yxEI14VpcEo1Z++WKE9HD7gUVFP7qgq5BWlwShVgBUrtIdzITjsBd5anoU4USGvKA1GqQKsWKHdNquVWy6ZQWtLEgFaW5LccsmM0M0dUUT61vIsxIlmoVSUBqNUf/pSvHei9gG3+hP2Au+SeVPzbPJQeXfUUlEhrygNRjkCbDiEdrFEYVoZDnfU4UKFvKI0GPUkwCC6SN9qfKCVggp5RWlA6kWAQX2ZVqJAhbyiKDVNvc1MwkaFvKIoNU89zUzCRl0oFUVR6hgV8oqiKHWMCnlFUZQ6RoW8oihKHaMLr4rS4NRLSl3FHRXyitLA1FNKXcUdNdcoSgNTTyl1FXdUk1eUBiaqlLpqAqoeVMgrSgMzVN6XUoS1moCqCzXXKEoD45cjvtQ87fVkAgq7GEklUCGvKA2MX2GPUoV1vVRViqIYSSVQc42iNDhueV86ulKuZhwYWliHmfq3krb9KIqRVALV5BVFycPSYL0YSliHVdu10pp0vcxIyhbyIjJKRJ4Wka0isl1EVma3TxGRp0TkeRFpF5ER5XdXUZSocdNgLYII67Bqu1batl8vdV7DMNccBM4xxrwtIgng/4jIT4CvALcZY+4TkX8FPgfcHsL1FEUJGbtZxPjsF1RYh5H6t9KadL0UIylbyBtjDPB29m0i+2eAc4DPZLffBaxAhbxSJ9STH7jT5dGL1pbksI7Ry7YfE2HK0vWRf+71UowklIVXEYkDW4CTgX8G/gT0GGP6s7vsAVw/GRFZBCwCmDx5chjdUZRIqTc/cD/zjEUlNFg3TRpgwGTmGsPxuddDMZJQFl6NMQPGmJnAROAMYFoRx95hjJltjJk9YcKEMLqjKJFSaVtxsQzl6+1n/hjKph6lH7nTth8XKdinmj/3aiFUF0pjTI+IbALOAlpEpCmrzU8Easu5VFE8qLStuBiCzDq8zCKtLUmeXHpOUede3N5N56793Ng2I5T+2zXpKUvXu+6jKRj8CcO7ZoKItGRfJ4G/Bp4DNgELsrtdBTxc7rUUpRrw8q4Yk0wMc0+GJsiso1SXR7dzG+CezS9H4uZYirfLUDONSrtpDgdhmGuOBTaJyLPAb4GfG2MeBb4OfEVEngeOBn4YwrUUpeIsmTeVRKzQdPDOof6ccKiWcPggs45SXB79gqUMRGJCKfZhFESA15rprRTC8K55Fpjlsv0FMvZ5Rakr2ma1svKR7RzoTedtTw+YnHColoXZoNGnQRcYO7pSrFi3nZ6+tO9+UZiuivV2CRKxWkumt1LRtAaKUgI9ve5C7pWevqoKhy/W19tunx6TTCCSGetxLUnOnjaBtVtSQ3riQP5DJEybdzHeLkEEeJgpGKoVTWugKCXgZx8ebu3QzzRUjCnGad7o6UtzoDedM3XcvfnlQALe/hCppM07iA0/rBQM1YwY4xffNrzMnj3bdHZ2VrobijIkbgFEyUScWy6ZweoNO0vyVgm7H8Vqy3NXPeFpZw9Kq0NT9zpnXIRBYyL1Zgn62QSdaVSzF46IbDHGzHZrU3ONopSA3T6c6ukjLpIzybiZNaLSDsM0DZUz0/B6sHidczgCmoLa8IOYgGo5AE6FvKKUiPXjdv74125JcenprWzasTdyrc9LiKZ6+pi76omirutlnx6Ksc0Jls+f7nqdIOf0eyiVqz2HFbFaTessxaJCXlFKwBI+bgKsLz3Aph17eXLpObn9Frd3s3rDzlCFfUdXiphITit2Uqy2uWTeVJY8uJX0QHATbksyQde3zvc9Z5C8ONbDyrnw+86h/lx/Kqk917IXji68KkqR2BcTvXilpy/SRUfr3F4C3sLP59u5YNu5az9eKShbkgkS8cLYABE8x2MJ7KDeOG4Lv84HTqV82Gs57bAKeUUpkiCCa0wyEWmgTVDhCYdNN3bvG7cH0D2bXyY9WCjlW1uSdC8/n9ULTqXFEdV7oDft+uAK8iC0sNYrgo6pEtpzLXvhqLlGUYokiJARiXaKX8w5BHLC1ppNjErEXFMS+F3LqvvqDISyP7gsU4ufGclOXCS3YLu4vTvQeCqhPddy2mEV8krdEpXLW5DFRCuAKKpAG69zC/nC2vkeMkI56CzAupaF30Kv3fYeRMA7PXKCfK6V1J5rNe2wmmuUuiRKe7jb1N2J9VCJaorvde4r5kzOC3wqNwrG6q9lv/c6n+VCOhRxEc+gLLcxJWLC2OZE7phLT8/MJiqdE6iWUE1eqUuidHmzjr/u/q2uGqtA3qyh1NmE10zEvqAZz5pFnEFIFl7BSMlEjL70oOe1BXKpDIbKVZNMxAPPDI5KNnm6Ww71eXV0pVjywNbcukGqp48lD2zNO1YpRCNelbpkytL1rlqnAC+uujDSawC8VOY13KI1LdOLmwnGy1f9ho5t3LP55bz9E7HMg8FljRXIeNJ0Lz8/cFlAP9z6WmpE7syVP3N92Fj9DZtqjnB1ohGvSkPh5z8e5qKdX7ENr36Vk0HROP63Y3m5AHma79otqQIb/YimGO8c8hbcVgGmYjx43Egm4oxKxAqydfalB7ju/q2590E/E6/ZxFAZMUuhliNcnagmr0RGJTQhP+3T0ipbHNkVS+3XULlR/AJ7nPs68Zsl+CHZQR7XkqT3UH+BgA10DjKznZL7wOE1icXt3Z7nSMQEhMCfyQkelaEg82AN8z7zMnNFkX8oDPw0eV14VSKhUtkH/bRPS5Q4sysueWBrSf3yy/BYbmBPqTMOY8iNqxQBb792S3Npla5eXHUhTy49h7ZZrb7jSA+aoj6TsT79Cfs+q+UIVycq5JVIqFTFnVJ+hOlBw4p120u6XtusVp5cek6eYIPgpg6v/gbx4ImCRFxy3jRvv9tf9PHOYtuljMPrM1k+f7pr1K2TMO6zWo5wdaJCXomESmlCpWqfYdt1g47TS2jYZwmQMYFEzdjmBKsXnJoLenKLfk3EvNccAD595qS899Y4nMLfD7/PZPWCU/NmTl6Ue5/VcoSrE114VSKhEhV3StU+oyBoRsfmETHmrnoiZ08+e9oE1+yVbusbbiUI3XDzcLG3uXkbeQnJ9OBh99AbOrZx71O7GTCGuAifPnMSN7bNKDjGLVsnuNvkBTh72gTPsTgDkrxs5+XeZ7Uc4epEhbwSCcWWnQsDL+0zCH723lIImtHxj6+/k3ttVV+yv7d7dDgFTOeu/Xn7e2Eg50/vxM8s4fWQWrFuO22zWrmxbYarUHfDS2h27tqf5+JpgLVbUsw+flzwzJk233nIPDzCuM9qNcLViQp5JRLK0YRK9copZ4p+4QePLflYT0JwXOtLD+SEqpNNO/YGPs+AMQVBS34P3SXzpnKtRy6Znr500bnqwV1ort6w0zXtgteYXXFagobDtlVDqJBXIqMUTagc/+RSi15AcQLT6qffg6icWYWTnr40HV2pwFWX3LAiYoM+PNtmtXoKeSjue3F+VnaTlNcn5DVmJ6s37CyYLaUHMgvp9WBqCYOyhbyITAL+AziGjO5yhzHmeyIyDmgHTgBeAi4zxhwo93pKfVNOOoKgBSrcGEpgFlvMIuwFZrfxB32oWRp7sQ/dsc0JX5u/U+N2e/BBYeWsICYmcB+zE6/PuacvnVtMr+VApjAIw7umH7jOGPMBYA7wdyLyAWAp8Lgx5n3A49n3iuKLl9AKIjSH8khJxLzn8X4LdaX4vIe9wOw2/iDuifZUvsUSxGXR0ri94iJWPrK95KhZtzE7C50E9aaqVLGRaqBsTd4Y8yrwavb1X0TkOaAVuBj4aHa3u4BfAF8v93pK/dLRlfL0BAkqNO3aqr1EX1zE13zSe6ifKUvXe5peivV5d5tVWGPzWgT1IyZS0D+rj35mlUFjfDVt+zjd2lcvONWzzKGFJTzdZmDlpEVwfuduprxETEjEJVDJwloMZAqDUNMaiMgJwK+A/wK8bIxpyW4X4ID13nHMImARwOTJk0/ftWtXaP1RagsvdzgBbls4M7TUA0ORiAmrP3Vq7npBw/udIe9OE4+VSiGM9L927dwv3N/q19nTJrB2S8r1oePV7kzR4PcwCRu39AZe90dLMsHokU25h5NXOocwUxJUW/KyYUlrICJHAGuBa40xb9nbTOZJ4npvG2PuMMbMNsbMnjDB2z9WqX+8NC1D6bbUUpJsOSNgg8wi3DxVrGjY2xbO5GD/YC6VQrkUa3qwSvt5JTyz7ORumvh192/NLYB6uZn6GXRakonAEa/WedxyzYP3/fFmXzov6nj5/OmRBjJVKmVHqYTiXSMiCTIC/h5jzEPZzX8WkWONMa+KyLHA62FcS6lfosjqWOoU3R4B6+aLHQPGNCd8k5zZzUVhYx/XUAukULo354AxuUXL5fOne5qg3Egm4qz4xHTAO/e+s49+2nbQALsgeenL0cKjrFUQBWVr8llTzA+B54wx/2hrWgdclX19FfBwuddS6ptiQsmDalPlLIDmncuhrsbjwoUfPJbjstkPV2/Ymbd/MYWsS8Ea13BE+doFmDMhm5/YtrTxtlmtDAY0C/s9lIu5P7xyCoWhhdda8rIwzDVzgc8C54hId/bv48Aq4K9F5I/Aedn3iuKJX1ZHJ0EToJWT6MtenNrNF/uezS97Cotyc7H7YRdsYfrj+2Ev5m0Xnl6zrNaWZN73FvRh67dfMfeHF2Ekzqu15GVheNf8H7zNcueWe36lehiOxaagvtx+BaXtQTT2qXuxWrV1Db+1Ajt2jTcqrc6qpWoJpeHSHr0EWND0FUFiGITM9+cXTVtuqoEwtPBKpOwoB414VQIRdqUcvyjIIA8Qv0Cgxe3ddO7an8urYgkGL+8Mv2v4VZlywxJSTbFMMq+wsfqR6unj2vZuYpLJIR8lfgLM+RB1PoSKedjaF4KjCl4KI3FerSUv08pQSiDCrJQTxK3Rq0KQfTFzqOyKTrdLP1fIREwKFlZHJWL0RiGpa4iYwKDBs1C4xVBVsuyctOyxQA/NKKowFdPPWkJrvCplM9Q0t9z6pU7cvBWcP1A/MWHIeHQsbu/O9cdLi2vJpimwMwgNL+CBXLHvobTrYjxOgs6KwjRFOWMWRiViZZd/rBVUyCuB8JvmFmvKCfrjde5X7GKm3bSxuL2bD500jv3vHCrQ4sSR01xxx16AO6irqtv2oBG/5XpGeeUa6ulLk0zESw6wqzW0MpQSCD/3tWI9FoLmG3H+yMvR7Azw6z/t59LTWwu8M3pKrIXaiFh+80530ZhH5Sc3Qe2sHuVGOQuZ5dbXrTdUk1cC4bfYtNgj3P2VrKeLc4E1iF+3W4WgclIJQ0bQP7r1VbqXn5+3PaqApbBJJmKMbIqHXqqwWOxmGEugumnmXoLaWhC3V5Wac+JYXnqjL5SFzHLr69YbKuSVsmnxiLhMJmJ5kaJ+aWYTDm8UA7T/dndehaByUglbWFkT4fADa1SiNia076YHufT0iYFT9UaJJSC9skwOlf2ymKpSpfZtKKrVrz1sVMgrgfCzu3uZV4tZuHTbNT1gWPnI9qLc8IJwbXt3XubCvhpZYB2TTHBPFQh4OLwW45VOYcCW/XK4CTLjq2a/9rCpDRVGqTh+dvc3IzQfOIWIFXFZLpVaaI172K6D0NNXXoKzZCLOmoUzyzjDYaxC4l64jdOZCz6qhF5u60eJmDC2OVFypGwto5p8gxDUxdFrPz/viXJt5UH6vuyhZ2tG4/ZjZJPQmx7+B4zdz73clMEimSLifknRnDb6sIPp/HBbP7KC7RpxkV2DoRqAoAEgfvt5mUgs4eHM0qhUB2ObEyyfPz0vQVdQId+STHCwf9DV5u4XiAYZTf7Wy07NLc4uvr/b1awXRcCTk3oNgLLjFwylQr4BCBqt6leUASjw6kjEhCNGNQ2Z5lapLNb31NObDpyiwRKCECxNsNc5Lj29lfand3sqAAK8uOrCos9dDGFGa1crtRPxunMnfPSj+dsuuwy+8AXo7YWPf7zwmKuvzvzt2wcLFhS2X3MNLFwIu3fDZz9b2H7ddTB/fuban/98YfsNN8B550F3N1x7bWH7zTfDhz4Ev/41XH99YfuaNTBzJmzcCDfeWNj+gx/A1KnwyCNw662F7T/6EUyaBO3tcPvthe0PPgjjx8Odd2b+nDz2GK/09HHlM+u5aMf/Lmxfmpky/27xN7l1bUde07tNI7n6spX09KX50pP3MnfX1lxbTIT9ySP5H22ZMX/tl3dyWmpH3vGvHjmexfO/CsC3Nt7BB15/Ia/9hXGtXH/BlwC4+af/xIn78220v3/Pifz9eYsAuO2R73LsX/bltT/TOo3vfORqAG7/8c2M7curVcOTx5/KP839NAB33r+cUf0H89ofP+kM/u3MSwC47z8LSxA/Ou2vuPu0CxmVfpc7H1hR0P7gjPN4cMZ5jO19k9s7bilov3vWx3n0/R/m2Lf2ctujhd/tv53xSR4/+UxOfGMPN2/4fkH7P33ocn4zZSbTXnuBbz1+R0H7dz58Fc9MfD+n7XmOr/3qroL2vz93Eb8/5kTOeKGLL/36voL26+d9kReOnsi5zz/F3z7949z2EU1xJo9LMn7R/TBpEhtX/E+u7Hqs4Phr2pZxoHkMC7ZtZMG2jQXtV39qBfc+tZtPb3nU9d67/DOrMh4u3/0uPPpofmMyCT/5Seb1P/wDPP54fvvRR8PatZnXy5bBb36T3z5xItx9NwCfe/B7rvfeN7L3HosWwR/+kH/8zJmZ3y7AlVfCnj357WedBbdkv/NLL4U33shvP/dc+OY3M68/9jHoczxkLroIvpr5bRTIPChf7tmoLiGvRIKfq9gJS9cjwKJnX+HsIs45aEzkibGUw2kFhoNjjhrFlPGjC7aPGz2i5HMONQNYMm8qPP5MyecPwhEj3cVco7hQqrmmAejoSrHkwa0auq+4IsAVcyZ7+q1HWd91TQmpBYpNed3oNnnV5BsFle+KC6NHxLnpk/7CrnPX/siuX2zJvFK8dGotNXDYqJBvAIarepBSfbQ4knM56T3kHz18Q8e2wBG2yUSMcaNHFuVOW2xqgVLrqxZTbGQ4iuMMJxoMVed0dKVqIi+LEj6tLUm6l5/P6gWneu5jwDNRV0dXKnCErZCJHH7nYD+JePCAr2Lt4lHXVw2jBmy1oUK+jrFuWKUxSfX0MXPlz7j+oWeH3M8tAnXlI9t9rXwxmyy39nPL+OhFKakFoq6vGkYN2GpDzTV1SjHTbKV+CZqx0m7bBlixbvuQx45sijEqEQ8cJ2H31y/VDBJ1fdWoZwqVQIV8HaICXimFvvQASx7oDlybti89GDjVRFyE1Z/KmI0se7ezDmwQol5EDaMGbLWhQr6OsNc/VZRSiCI9kOWu2LlrP/dsfrnsgt3ORVQr8VkYQj/qmUIlCMUmLyL/LiKvi8jvbNvGicjPReSP2f/HhnEtxR37gpGiDEXpuTAPM7Y5kUt5UXB+IS/jI5An4C3KtXeHvVDaNquVWy6ZUVA9rJa9a8LS5O8Evg/8h23bUuBxY8wqEVmaff/1kK6nOCi2/qnS2HzopHE8+afS/d8TcWH5/Ol07tpfYBpMxDKmGbtgnLvqCc9F3HLs3aW6VPpRjLtlLRCKkDfG/EpETnBsvhj4aPb1XcAvUCEfKnZ/XvWCV4qhHAEvwML/mqnT2v7b3QXtC8+YVCAk/WaY5di763GhNGyidKE8xhjzavb1a8AxbjuJyCIR6RSRzr1790bYnfrCOU1VlOHCAJt27GXlI9td3SXXP/tqwTa/Yinl2LujdqmsB4bFT95kEuS4yiJjzB3GmNnGmNkTJkxw20WxYS0yXdvereYZpWK80tPn6Trptt0vUVk5phG3KlC1vlAaNlEK+T+LyLEA2f9fj/BaDYEuripuhLGIWizFzh5bPTRrr+1BqceF0rCJ0oVyHXAVsCr7/8MRXqsh0MVVxY0ozXXJRIx304NFXcPN4yZK18R6WygNm1CEvIjcS2aRdbyI7AGWkxHu94vI54BdwGVhXKvR0MVVpZIcGjBF3XeJmLDiE9Nd20Y2xXJC3lmWsBao1cRlYXnXfNqj6dwwzt+ouOXBVpThZKCI7KWtHoLP7T5+t8aKsg9nIfKw0YjXKkQjV5UwGKrYdpj41UuNwpd9uKnlMaiQrzJUe1fCYjjNe35+6fXgy17LY1AhHxGl2u90cVUJm6AafTIRL/neO64l6XnP10PSr1oeg+aTj4By8mnUgmag1BYGd48XJ7dcMgOfmCVPkok4Z0+b4HnP14Mvey2PQYV8BJRTeKAWNAOltmhJJnjnYP+Q+7XNamXMqKEfBnYsv/RNO/a63vMr1m2vC1/2Wh6DmmsioBz73dnTJrhm61OUUkjEhLfeTTOUk8zoERkt9c2ARUYgf7F1cXu36z49fWk6ulJ14cteq2NQTT4CSs2n0dGVov3p3SrglVCQ7D9BvCAT8YwoKGYmaVda/I6r5dJ59YAK+Qgo1X63Yt120kX4JSuNQympCwwErrdqafBu964XY2x2fr97W9eZKosK+Qgo1X4XtB6n0nhE/ei3e8f0pQdyWSNbW5JcOWcyiVjhY+adQ/05Z4K2Wa2MbXa35+s6U2VRm3xEBLXfaeCTUmns3jHW4umAMbnZZ9usVtY/+2pBdsn0gMkLBlo+f3rdlc6rB1TIV4iMm+WzgQshK0qYtLYk8/zZVz6y3Teis8cjrbDdFBN1kW2lNFTIDxMdXSlWrNuuJhmlLErJCunEmYLgho5tnrnhLSEeNBioVj1Q6hm1yQ8DHV0pljywVQW8UjanTW6hqYxfrdN80tGV4h5HjVY7lhCv5WCgRkc1+WFg9Yad6jWjhEKptVkF8swnQdeCLCE+lCmmVtPwNgIq5CNC88Ar1YLTPBM0CV5LMpEnqL1MMbWchrcRUCEfEnahPiaZ4J1D/YF9lBUlKhIxKTCpBEmCJ+BZ/MNJLafhbQRUyIeAU5NR27tSLaQHDSsf2Q4c1qqHCk4S4Io5k/PMOn5mmFpOw9sIqJAPAU0PrFQzB3rTeeYTL08ZyK/uFNQMU8tpeBsB9a4JAdVYlGrHngXVy1NmzcKZPLn0nLxF1iDZVNXzprpRTT4E/DQjRYmSuAgDJtjaj6WMBA1aCmqG0SCo6kaFfIlocJNSaSyvmSlL1wfy4LKbT4IELRVjhtEgqOpFzTU+dHSlmLvqCaYsXc/cVU/kkjFpcJNSDaR6+pi76glaPBKD2RH8M0W6oWaY+iByTV5ELgC+B8SB/2WMWRX1NcPAb9Fp5SOaElipDlI9fSRiQiIuvi67huJ91tUMUx9EKuRFJA78M/DXwB7gtyKyzhjz+yivGwZei06aVEypNtKDhpZkgtEjmzzXhuKlFG9FzTD1QNSa/BnA88aYFwBE5D7gYsBVyHd399LSkl9GbMKETbS2PszAwEi2bft2wTHvfe9Pee97f0o6PYbt21cWtB933MO85z2bePfdCezY8Y2C9okT2xk//jf09k7iD3+4Lrf9rXePBI6k6ag7iI3azOChqfT3fL3g+KYx3yM2ciuDB0+l/80vF7a3fJvYiJ0MvjuH/rcWFbaP/XtiiZcY6PsIA3+5qqA9MW4Z0vRnBnrnMfD2wsL2o7+CxHsYeOdiBt65uLB9/BeQ2LsM/GUhA33zCtpHvOdvAOh/6yoG3/1IfqMcZMSEazLtb36ewYNn5jfHekiM/0qmvefLDB46Nb89/mcSRy/LtB/4GoPpafntTbtIjMt8Z+n9yzH9x+e1xxI7aBr7nUz7G7dgBo7Jbx+xlaaW72Xa9/0jZrAlv33kUzSN+QEAh/beDmZkfvuoX9J01F2Z9tf/HSfx5AbiR7ZjBkeR3vcvhe2jHyY++mHMQAvpN/6xsP2IduLNGzD9x5Def0th+5F3EU/+ksH0CfQf+FZBezH33stvfpkR8RiHBo483O6491puz/9tnXLKrTQ372bfvrPYs6fw3po27SZGjdrL66+fzSuvFN5b06cvJ5F4k9deu4DXXrugoH3GjNQ6ZfcAABZhSURBVK8Tjx8klbqYvXvPLmifOfNaAHbvXsgbb5yV1xaLHeSDH8yMedeuz3LgwOl57YnEm0yfvhyAF174W956Kz9wa+TIvbz//TcB8PzzX+Ttt0/Oa29u3s0pp9wKwB/+cB29vZPy2o844nlOPvn7ADz33Dc4eHBCXvtRR23nxBP/DYDt21eSTo/Jax87dgvHH/8jAJ599tsMDubfe0cf/RsmTWoHoLt7DU7KlXt2orbJtwK7be/3ZLflEJFFItIpIp2DVWQCiZWo+ShKJRDg0ID3DFPv58ZFTED3q5JOLrIAuMAY89+z7z8LnGmM+aLb/rNnzzadnZ2R9acYgub3UJRKk0zEiQm8c8j7Xl2zcKaaXeoYEdlijJnt1ha1Jp8C7POgidltVY9bCT+v8maKUgnspSX9BLwz0ZjSWERtk/8t8D4RmUJGuF8OfCbia4aGc9FJtXulWnBmlry2vdtz34tOPXY4uqRUKZFq8saYfuCLwAbgOeB+Y8z2KK8ZJZ279quAVyqOm696MuH9U167JZWL8VAaj8iDoYwxjxljTjHGnGSMuSnq60XFDR3buNungo6iDAcCXHp6oVvjKEfQkh23fDNK46ARrwG596ndQ++kKBFjgE079hZs9yq0baFJ9BoXzV2TxZk3++xpE9i0Yy+pnr6ikkApStS4CeyhkuRp2t/GRYU87ikM7KYZFfBKVCQTcW65ZAZAXmUxkUweeDcsge2sRuaV2kDzzTQ2aq5Bi34olcNeJm/JvKkc15Lkzb40zSOauHLOZM8EYZZiksrWEO7pS4Mh5+ZrpTGwXCzVhbJxUU0etVcqlSXV05fLbGolvkv19NH+9G4WnjGJTTv2FiQIm7vqiQLFJD1oaB7RRNe3zq/EMJQqRYU8WvRDCQ+BQLndncdc/9CzBZlN04OGR7e+SvfyQqGtdVWVoKi5Bve82YpSDAK8tOpCrpgzuehjDdDrkdnUq2aB10KqLrAqTlTIcziFQanpWBXFAHNXPcEDnXuG5Xpa0EMJigr5LG2zWhlULxqlDFI9fRzsD7fWgFe+JLfcSrrAqrihNnkbaptXoiYGFPMYWD5/umebV0EPZ8yHVnNqbBpOyDt9iw/1D+TsoTqtUaJEgOSIuG/GSDulZI/0K1upgr4xaSi55uZbbF/w0qJ+SpQY/HO+20km4qz4hLcW74VX2UrNXdO4NJQmr0FPSrVi1Wgt18SirpWKk4bQ5Du6Usxd9YTa25WK05JMFHjFJGLCof6B3Ayz91B/yedX10rFSd0LebuJRlGGg7HNCdYsnOnq4rjiE9PzvGJakgkGjMkzGx7oTbPkwa1F54Dv6ErxzsHCB4S6VjY2dS/k1USjDDcHetOs3rCTS09vdXVxbJvVypNLz+HFVRcyemQTbvXr0wOmKDu6pcw4g6fGNifUtbLBqXubvNoilahpzbre2lMapHr6WLslNaSA9bs/i7l3vZSZ5hFNKuAbnLrX5NUWqUSJVWu1tSVZkLPG7tVirQtNWbqeuaueyJli/O7PYu5dXXBVvKh7IX/2tAmV7oJSp9ht3X5C1um6a/mud3SlWDJvKolYYTqNRFyKsqPrgqviRd0LebdSaYpSLs40An5C1s93vW1WK6s/dSotycPpC8Y2J1i94NSizCyay0bxQm3yiuJDIiYcMaqJnt60r//6knlT8yJN4bCQXdze7Xpu6970Sk9QDNbxms5AcVL3Qj7KfDRxySy0uXlHKLVPaxGC0k3Inj1tAqs37PTMLx+2KSWMh4VSf5RlrhGRT4nIdhEZFJHZjrZlIvK8iOwUkXnldbN0orTJDxgV8PXMKz19rN6wM7C/ut01csm8qazdkvJUMNSUogwX5WryvwMuAX5g3ygiHwAuB6YDxwEbReQUY8ywO6yvf/bV4b6kUidYi6RfyZpbitGS/eIzipkhKEq5lKXJG2OeM8a4RWxcDNxnjDlojHkReB44o5xrlUJHV8qz4n1QgpQR8aoqlUzEPfOBK7XDILDsoWeLOsZrLUiAJ5eeowJeGTai8q5pBXbb3u/JbitARBaJSKeIdO7dG64nTBiZ966YMznP88FJXCQXpm69h8PeF8vnT9fSgnVAn0d5PnD3gVeXRqVaGNJcIyIbgfe6NH3DGPNwuR0wxtwB3AEwe/bsUC3c5XrWJBMxbmybwezjx3Gth4fEp8+cVLDgZeWsX9zezXEtSS49vZVNO/Zq/pwqZWxzIuc9U8x31NGVYsW67XmpBCwf+EtPb2XtlpSrt42iDCdDCnljzHklnDcFTLK9n5jdNqyU61ljaW+WAF/20LO5bTGBz5w5mRvbZhQUInnnUD/pgczzyh7e/kDnyzz5p/1ljkoJm65vnZ97feKy9a6L6c54JWdxDjt96QE27djLLZfMUJdGpeJE5UK5DvhPEflHMguv7wOejuhaniyZN9VTAy8Wv1Jr9h+7M0EUZH70192/lQGtIVt1tDrMJ2edOM71QfyZMyfnvR8q8d0rPX3q0qhUBeW6UH5SRPYAZwHrRWQDgDFmO3A/8Hvgp8DfVcKzpm1W65ALn2ObE4we4W4zH+rYGzq2cW17d6Aslyrgqw+n+aSjK8UzL79ZsN/ck8ZxY9uMvG1DmQLV9q5UC2Vp8saYHwM/9mi7CbipnPOHwfL5010jEe0h6R1dKZY8uDVnYoFM7hC/Iso3dGzj7s0vR9dxpWRikolfaHHU8LUztjnB8vnTaZvVmjO3eZn2XnqjcLufKVBt70o1Ufe5a9pmteYVaXDmHLH2Wb3g1Lx9hsodcu9Tuz3blMoysinOmoUz6V5+PmNHj3Tdx0rBe0PHNha3d/uu3bhp7W65YkDztyvVR92kNbAvfjoXuZwh55ZrpVPQF/PDVPNL9WJP/uUlvFPZ7JD3bH7ZM+2AhZvpRXPFKLVCXQh55+Kn5cYG5Kbjfu2lEBcZUtAnE3FOmzxGPWoqgKV9e31PcRHfvDIWfqYXXVhVaoG6MNd4pXJdsW47c1c94bo4ai/oUAqfPnOS6/bRI+J5ZiE3e64SPZb27fUgHjBmyMVTN9OeotQadaHJe/1Ye/rSri6NQx0XhNnHj2Ptlj0FkZAtzSO46ZOHp+1eaWaV6BDIad+tHgukluukW5sAty2cqcJdqQuqS8jv3Akf/Wj+tssugy98AXp74eMfLzzm6qs5rmUyva+8xu0dtxQ03z3r4zz6/g9z7Ft7ue3RW/PaRjTFYe4gzJ+fufbnP194/htugPPOg+5uuPZa9r19kJf29fLewUH+f+A7H76KZya+n9P2PMfXfnVX5ph/gRePGsWU8aM5Y9pCnmqZzNyXuvnSr+8rOP31877IC0dP5Nznn+Jvny50VFp80XW8etQELnruV1zZ9VhB+zVtyzjQPIYF2zayYNvGwo/nUyt4NzGKK59Zz0U7/ndB++WfWQXA3z71EOf+KT+U4d2mkVx92UoAvvTkvczdtTWv/UDyKK755PUAfO2Xd3Jaakde+6tHjmfx/K8C8K2Nd/CB11/Ia39hXCvXX/AlAG7+6T9x4v78eLnfv+dE/v68RQDc9sh3OfYv+3JtsZjQeexUVn/kagxw+49vZmzfWwAcc9QopmweDeeey5KL/oZlD23j9ntuYFT/wcyxIpw4YTSv/dV5fOqIufSlB7jvP5fmzp07PsC9x9VXw759sGBBYfs118DChbB7N3z2s4Xt111X1L1XwM03w4c+BL/+NVx/fWH7mjUwcyZs3Ag33ljY/oMfwNSp8MgjcOuthe0/+hFMmgTt7XD77YXtDz4I48fDnXdm/pw89hg0N8O//Avcf39h+y9+kfn/u9+FRx/Nb0sm4Sc/ybz+h3+Axx/Pbz/6aFi7NvN62TL4zW/y2ydOhLvvzry+9trMZ2jnlFPgjjsyrxctgj/8Ib995szM5wdw5ZWwZ09++1lnwS1ZeXPppfDGG/nt554L3/xm5vXHPgZ9DmXioovgq5nfRoHMg/LvPRt1Ya5ZMm8qo4rMDxMTYfI4d1/mfW8f5JmXe9j8wht8+b6uXKrZfW8f5IW979A/6J3HxOLPb73LvrcPBstwphTNGSeM4wsfPZnbFs7MaeUjmuKc/J4jmDJ+NADPvfqWqykvlg1f/S+tY/LyDjmPV5R6QEwVeYnMnj3bdHZ2lnSsM7WAn5nGL9WrW7i65Vfv50vtdZ1XsnU9gzK2OVF25sxaRwQwmQexm03dKp7th1/aASiMlVCUWkZEthhjZru11YUmD/kFG0aPdLdCCbBm4UzfVK9+9TiLteFbrnXFcOEHjy1q/3pDgNsum8mLqy7k1stOLblu6VBpB8pdeFeUWqFuhLwdL2FsGNpl0uvYUgS25TsdNNVwXKTho2jt31GQQDYvgjyQtf6v0ghU18JrSHiFnDuTURVzrCWw3UwAiRg4I+ctjdMSSM6UtG5ogFXmQTdl6fpcjdRNO/aWFGwUJAOp5pdRGoG61OTdtOeg03y/Y900yzULZ/LHmy9kTXYB0E3jbJvVSvfy8/P2aUkmGNucQGjctVm3cQ8Ykyu7d/fml0ll1zSsALag9VaHmkFpfhmlUaibhVcnfmkOojy2WBo50dmahTNzn7PXIquTIIuuFs7FeBFyxUE0BYFST/gtvNa8kI9SIA+HsD9p2WMNaaZxBhxNWbo+kBeSAC+uujDKrilKzVG33jWWm1ypU/pKndtOKQI+bPNOXIS5J40L+az+GPJr8Aa1j6sdXVGKo6aFvJ+7YzWf245V+DsorS1Jbls4MzRB35JMcFSyiV8HTKKWCPGOsXu3BPFCUju6ohRPTQt5P3fHqM6d6uljytL1zF31RChavVeiMyetLUleWnVhzsd/KI026EOgpy/Ngd504IAtl/obJWMfg7Wo7fXQi4to8JKilEBNC3kvQRfGlN7vHGGab25sm8GVcyb7CmU3DdZN87XOYWn7QVxGwyaot5DbmNpmtTLoYb4aNEYFvKKUQE0L+XJcJUs5t5OwzDc3ts3wfKh4abBts1q59PTWnOYbF+GKOZPztP1iArHCoLUlyYurLuS2hTNJxPJFfQxyLqOtLUkuPb2V1Rt2FsyKonxwK0ojUtPBUFFW53Ge28ucEVbUpNd5vDTYjq4Ua7ekcgu3A8awdkuK2ceP86yIdVxLkt5D/UPmxrG7Kc5c+bMhg7ggP73vUN+LXxEXt4AztcUrSunUtJCHaKvz2M89d9UTnpGwYeAXaeuG38KwX1lDt6LlTnoP9dPRlaJtVitB1oUFuGLO5MDlFP36bj1ctKyeooRDzQv54SJqDbPY85e16DzEKuuB3nROs+7x0foFShLCQ/Vdy+opSniUJeRFZDUwHzgE/An4b8aYnmzbMuBzwADw/xljNpTZ14oSdeHmYs9frOZvsXrDTtKDQ/vSWJq1Xx6goJGnbn2MclakKMphytXkfw4sM8b0i8i3gWXA10XkA8DlwHTgOGCjiJxijPHO/VoDRK1hFnP+UmcWxawhpHr6WLNwZugzGLW7K8rwUZaQN8b8zPZ2M2DVoboYuM8YcxB4UUSeB84AHDW6lFIpdWYRJDtjqdcJmgYi6lmRoiiHCS13jYg8ArQbY+4Wke8Dm40xd2fbfgj8xBjzoMtxi4BFAJMnTz59165dofRHyeAUvGdPm8DaLSnfghp2XgqYJ8avopYKb0WJlrJy14jIRhH5ncvfxbZ9vgH0A/cU2zljzB3GmNnGmNkTJkwo9nDFB7f8O2u3pLj09Na8tMgxDw+aYlIuDFcaCEVRimNIc40x5jy/dhG5GrgIONccnhakAHu8/sTsNmUY8RK8m3bszVs09Up3HDTlAkSbYkJRlNIpK+JVRC4AvgZ8whjTa2taB1wuIiNFZArwPuDpcq6lFE9QwWulVrBHz145ZzI3ts0IfC2NVFWU6qRc75rvAyOBn0tGQGw2xvwPY8x2Ebkf+D0ZM87f1bpnTS1SjKvijW0zihLqTtRjRlGqk3K9a072absJuKmc8yvlMZyCVz1mFKU60YjXOiYqwevlKqmRqopSfaiQr3PCFrx+ycVUwCtK9VHTqYaV4UddJRWltlAhrxSFukoqSm2hQl4pCnWVVJTaQoW8UhRRVuNSFCV8dOFVKQp1lVSU2kKFvFI06iqpKLWDmmsURVHqGBXyiqIodYwKeUVRlDpGhbyiKEodo0JeURSljgmt/F8YiMheoNbr/40H9lW6EyFRL2PRcVQf9TKWahnH8cYY19J6VSXk6wER6fSqtVhr1MtYdBzVR72MpRbGoeYaRVGUOkaFvKIoSh2jQj587qh0B0KkXsai46g+6mUsVT8OtckriqLUMarJK4qi1DEq5BVFUeoYFfJlIiKLRWS7iPxORO4VkVEiMkVEnhKR50WkXURGVLqfTkTk30XkdRH5nW3bOBH5uYj8Mfv/2Ox2EZH/mR3PsyJyWuV6no/HOFaLyI5sX38sIi22tmXZcewUkXmV6bU7bmOxtV0nIkZExmff19R3kt3+pez3sl1EvmPbXlPfiYjMFJHNItItIp0ickZ2e3V+J8YY/SvxD2gFXgSS2ff3A1dn/788u+1fgWsq3VeXvn8YOA34nW3bd4Cl2ddLgW9nX38c+AkgwBzgqUr3f4hxnA80ZV9/2zaODwBbgZHAFOBPQLzSY/AbS3b7JGADmUDB8TX6nZwNbARGZt+/p1a/E+BnwMds38Mvqvk7UU2+fJqApIg0Ac3Aq8A5wIPZ9ruAtgr1zRNjzK+A/Y7NF5PpL+T3+2LgP0yGzUCLiBw7PD31x20cxpifGWP6s283AxOzry8G7jPGHDTGvAg8D5wxbJ0dAo/vBOA24GuA3Uuipr4T4BpglTHmYHaf17Pba/E7McBR2ddjgFeyr6vyO1EhXwbGmBTwXeBlMsL9TWAL0GMTMnvIaPy1wDHGmFezr18Djsm+bgV22/arpTH9DRntCmpwHCJyMZAyxmx1NNXaWE4B/iprxvyliPzX7PZaGwfAtcBqEdlN5ve/LLu9KseiQr4Msjbri8lMM48DRgMXVLRTIWEy88+a9q8VkW8A/cA9le5LKYhIM3A98K1K9yUEmoBxZMwYS4D7RUQq26WSuQZYbIyZBCwGfljh/viiQr48zgNeNMbsNcakgYeAuWSmaVZpxYlAqlIdLJI/W9PL7P/WlDpFxi5sUfVjEpGrgYuAK7IPLKi9cZxERoHYKiIvkenvMyLyXmpvLHuAh7KmjKeBQTLJvWptHABXkfmtAzzAYfNSVY5FhXx5vAzMEZHmrFZyLvB7YBOwILvPVcDDFepfsawj01/I7/c64P/Neg/MAd60mXWqDhG5gIwN+xPGmF5b0zrgchEZKSJTgPcBT1eij0EwxmwzxrzHGHOCMeYEMoLyNGPMa9TYdwJ0kFl8RUROAUaQyd5YU99JlleAj2RfnwP8Mfu6Or+TSq/81vofsBLYAfwO+BEZL4ETydyoz5N50o+sdD9d+n0vmXWENBnh8TngaOBxMjftRmBcdl8B/pmM58M2YHal+z/EOJ4nYxvtzv79q23/b2THsZOsh0S1/LmNxdH+Eoe9a2rtOxkB3J39nTwDnFOr3wnw/5BZe9sKPAWcXs3fiaY1UBRFqWPUXKMoilLHqJBXFEWpY1TIK4qi1DEq5BVFUeoYFfKKoih1jAp5RVGUOkaFvKIoSh3zfwHuo5wwGN9VqQAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "LOAA_5: 27.99228751475208\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "adchbmhaehhc",
        "outputId": "5826f004-fc6a-4c8c-88bb-7181f8bfefb4"
      },
      "source": [
        "#Model 6\n",
        "X6 = df[['BMXWT', 'BMXWAIST']]\n",
        "\n",
        "model_6 = linear_model.LinearRegression()\n",
        "model_6.fit(X6, y)\n",
        "\n",
        "print(model_6.coef_)\n",
        "print(model_6.intercept_)\n",
        "\n",
        "#Add a constant:\n",
        "X6a = sm.add_constant(X6)\n",
        "model_6a = sm.OLS(y, X6a)\n",
        "\n",
        "results_model_6a = model_6a.fit()\n",
        "print(results_model_6a.summary())\n",
        "\n",
        "#Residual Standard Error (RSE):\n",
        "model_6a = sm.OLS(y, X6a).fit()\n",
        "model_6a.resid.std(ddof=X6a.shape[1])\n",
        "\n",
        "#Bland-Altman Plot:\n",
        "df['BMXHIP_6'] = 45.073412 + (0.307425*df['BMXWT']) + (0.365595*df['BMXWAIST'])\n",
        "df['md_6'] = df['BMXHIP']-df['BMXHIP_6']\n",
        "print(df.head(10))\n",
        "\n",
        "df['mean_BMXHIP_6'] = statistics.mean(df['BMXHIP_6'])\n",
        "df['sd_BMXHIP_6'] = statistics.stdev(df['BMXHIP_6'])\n",
        "df['mean_md_6'] = statistics.mean(df['md_6'])\n",
        "df['sd_md_6'] = statistics.stdev(df['md_6'])\n",
        "print(df.head(10))\n",
        "\n",
        "#Plot drawn:\n",
        "print(\"Mean md_6:\", statistics.mean(df['md_6']))\n",
        "print(\"SD md_6: \", statistics.stdev(df['md_6']))\n",
        "print(\"LLA: \", statistics.mean(df['md_6'])-(1.96*statistics.stdev(df['md_6'])))\n",
        "print(\"ULA: \", statistics.mean(df['md_6'])+(1.96*statistics.stdev(df['md_6'])))\n",
        "\n",
        "plt.plot( 'BMXHIP', 'md_6', data=df, linestyle='none', marker='o')\n",
        "plt.axhline(y = 0, color = 'black', linestyle = '-')\n",
        "plt.axhline(y = -2.641691317246042e-06, color = 'blue', linestyle = '--')\n",
        "plt.axhline(y = -12.79523587443649, color = 'red', linestyle = '--')\n",
        "plt.axhline(y = 12.795230591053855, color = 'red', linestyle = '--')\n",
        "plt.show()\n",
        "\n",
        "#Limits of Agreement amplitude:\n",
        "LOAA_6 = 12.79523587443649 + 12.795230591053855\n",
        "print(\"LOAA_6:\" , LOAA_6)"
      ],
      "execution_count": 52,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[0.30742543 0.36559462]\n",
            "45.073412271414156\n",
            "                            OLS Regression Results                            \n",
            "==============================================================================\n",
            "Dep. Variable:                 BMXHIP   R-squared:                       0.809\n",
            "Model:                            OLS   Adj. R-squared:                  0.809\n",
            "Method:                 Least Squares   F-statistic:                 2.079e+04\n",
            "Date:                Mon, 22 Nov 2021   Prob (F-statistic):               0.00\n",
            "Time:                        21:12:45   Log-Likelihood:                -32301.\n",
            "No. Observations:                9803   AIC:                         6.461e+04\n",
            "Df Residuals:                    9800   BIC:                         6.463e+04\n",
            "Df Model:                           2                                         \n",
            "Covariance Type:            nonrobust                                         \n",
            "==============================================================================\n",
            "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const         45.0734      0.413    109.033      0.000      44.263      45.884\n",
            "BMXWT          0.3074      0.007     46.186      0.000       0.294       0.320\n",
            "BMXWAIST       0.3656      0.008     43.469      0.000       0.349       0.382\n",
            "==============================================================================\n",
            "Omnibus:                      496.370   Durbin-Watson:                   0.883\n",
            "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              636.180\n",
            "Skew:                           0.507   Prob(JB):                    7.17e-139\n",
            "Kurtosis:                       3.726   Cond. No.                         815.\n",
            "==============================================================================\n",
            "\n",
            "Warnings:\n",
            "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
            "   BMXWT  BMXHT  BMXWAIST  BMXHIP  ...  mean_md_5  sd_md_5   BMXHIP_6       md_6\n",
            "0   57.9  168.2      68.0    66.6  ...   0.000018  7.14089  87.733779 -21.133780\n",
            "1   34.0  145.4      60.6    67.4  ...   0.000018  7.14089  77.680919 -10.280919\n",
            "2   52.5  169.3      68.0    68.7  ...   0.000018  7.14089  86.073684 -17.373684\n",
            "3   31.6  138.5      62.9    69.1  ...   0.000018  7.14089  77.783967  -8.683967\n",
            "4   30.7  140.5      59.2    69.7  ...   0.000018  7.14089  76.154584  -6.454583\n",
            "5   29.4  141.3      55.2    69.9  ...   0.000018  7.14089  74.292551  -4.392551\n",
            "6   36.5  150.5      59.8    69.9  ...   0.000018  7.14089  78.157005  -8.257005\n",
            "7   31.4  134.0      59.4    70.0  ...   0.000018  7.14089  76.442900  -6.442900\n",
            "8   29.2  137.5      56.4    70.1  ...   0.000018  7.14089  74.669780  -4.569780\n",
            "9   35.2  144.8      64.7    70.5  ...   0.000018  7.14089  79.548768  -9.048768\n",
            "\n",
            "[10 rows x 38 columns]\n",
            "   BMXWT  BMXHT  BMXWAIST  ...  sd_BMXHIP_6  mean_md_6  sd_md_6\n",
            "0   57.9  168.2      68.0  ...    13.447275  -0.000003  6.52818\n",
            "1   34.0  145.4      60.6  ...    13.447275  -0.000003  6.52818\n",
            "2   52.5  169.3      68.0  ...    13.447275  -0.000003  6.52818\n",
            "3   31.6  138.5      62.9  ...    13.447275  -0.000003  6.52818\n",
            "4   30.7  140.5      59.2  ...    13.447275  -0.000003  6.52818\n",
            "5   29.4  141.3      55.2  ...    13.447275  -0.000003  6.52818\n",
            "6   36.5  150.5      59.8  ...    13.447275  -0.000003  6.52818\n",
            "7   31.4  134.0      59.4  ...    13.447275  -0.000003  6.52818\n",
            "8   29.2  137.5      56.4  ...    13.447275  -0.000003  6.52818\n",
            "9   35.2  144.8      64.7  ...    13.447275  -0.000003  6.52818\n",
            "\n",
            "[10 rows x 42 columns]\n",
            "Mean md_6: -2.641691317246042e-06\n",
            "SD md_6:  6.528180220788353\n",
            "LLA:  -12.79523587443649\n",
            "ULA:  12.795230591053855\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "LOAA_6: 25.590466465490344\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "8cHe6qK2evm0",
        "outputId": "bf93fa97-3fbe-4c54-88b0-5bfd44de7b37"
      },
      "source": [
        "#Model 7\n",
        "X7 = df[['BMXHT','BMXWT', 'BMXWAIST']]\n",
        "\n",
        "model_7 = linear_model.LinearRegression()\n",
        "model_7.fit(X7, y)\n",
        "\n",
        "print(model_7.coef_)\n",
        "print(model_7.intercept_)\n",
        "\n",
        "#Add a constant:\n",
        "X7a = sm.add_constant(X7)\n",
        "model_7a = sm.OLS(y, X7a)\n",
        "\n",
        "results_model_7a = model_7a.fit()\n",
        "print(results_model_7a.summary())\n",
        "\n",
        "#Residual Standard Error\n",
        "model_7a = sm.OLS(y, X7a).fit()\n",
        "model_7a.resid.std(ddof=X7a.shape[1])\n",
        "\n",
        "#Bland_Altman Plot:\n",
        "df['BMXHIP_7'] = 130.824112 + (-0.488518*df['BMXHT']) + (0.617183*df['BMXWT']) + (0.064531*df['BMXWAIST'])\n",
        "df['md_7'] = df['BMXHIP']-df['BMXHIP_7']\n",
        "print(df.head(10))\n",
        "\n",
        "df['mean_BMXHIP_7'] = statistics.mean(df['BMXHIP_7'])\n",
        "df['sd_BMXHIP_7'] = statistics.stdev(df['BMXHIP_7'])\n",
        "df['mean_md_7'] = statistics.mean(df['md_7'])\n",
        "df['sd_md_7'] = statistics.stdev(df['md_7'])\n",
        "print(df.head(10))\n",
        "\n",
        "#Plot drawn:\n",
        "print(\"Mean md_7:\", statistics.mean(df['md_7']))\n",
        "print(\"SD md_7: \", statistics.stdev(df['md_7']))\n",
        "print(\"LLA: \", statistics.mean(df['md_7'])-(1.96*statistics.stdev(df['md_7'])))\n",
        "print(\"ULA: \", statistics.mean(df['md_7'])+(1.96*statistics.stdev(df['md_7'])))\n",
        "\n",
        "plt.plot( 'BMXHIP', 'md_7', data=df, linestyle='none', marker='o')\n",
        "plt.axhline(y = 0, color = 'black', linestyle = '-')\n",
        "plt.axhline(y = -4.144073241301843e-05, color = 'blue', linestyle = '--')\n",
        "plt.axhline(y = -10.497097910321404, color = 'red', linestyle = '--')\n",
        "plt.axhline(y = 10.497180791786231, color = 'red', linestyle = '--')\n",
        "plt.show()\n",
        "\n",
        "#Limits of Agreement amplitude:\n",
        "LOAA_7 = 10.497097910321404 + 10.497180791786231\n",
        "print(\"LOAA_7:\" , LOAA_7)"
      ],
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[-0.48851763  0.617183    0.0645308 ]\n",
            "130.8241119215765\n",
            "                            OLS Regression Results                            \n",
            "==============================================================================\n",
            "Dep. Variable:                 BMXHIP   R-squared:                       0.872\n",
            "Model:                            OLS   Adj. R-squared:                  0.872\n",
            "Method:                 Least Squares   F-statistic:                 2.218e+04\n",
            "Date:                Mon, 22 Nov 2021   Prob (F-statistic):               0.00\n",
            "Time:                        21:12:45   Log-Likelihood:                -30360.\n",
            "No. Observations:                9803   AIC:                         6.073e+04\n",
            "Df Residuals:                    9799   BIC:                         6.076e+04\n",
            "Df Model:                           3                                         \n",
            "Covariance Type:            nonrobust                                         \n",
            "==============================================================================\n",
            "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
            "------------------------------------------------------------------------------\n",
            "const        130.8241      1.288    101.546      0.000     128.299     133.349\n",
            "BMXHT         -0.4885      0.007    -68.994      0.000      -0.502      -0.475\n",
            "BMXWT          0.6172      0.007     87.300      0.000       0.603       0.631\n",
            "BMXWAIST       0.0645      0.008      7.904      0.000       0.049       0.081\n",
            "==============================================================================\n",
            "Omnibus:                      164.970   Durbin-Watson:                   1.212\n",
            "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              246.299\n",
            "Skew:                           0.186   Prob(JB):                     3.29e-54\n",
            "Kurtosis:                       3.681   Cond. No.                     5.01e+03\n",
            "==============================================================================\n",
            "\n",
            "Warnings:\n",
            "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
            "[2] The condition number is large, 5.01e+03. This might indicate that there are\n",
            "strong multicollinearity or other numerical problems.\n",
            "   BMXWT  BMXHT  BMXWAIST  BMXHIP  ...  mean_md_6  sd_md_6   BMXHIP_7       md_7\n",
            "0   57.9  168.2      68.0    66.6  ...  -0.000003  6.52818  88.778388 -22.178388\n",
            "1   34.0  145.4      60.6    67.4  ...  -0.000003  6.52818  84.688395 -17.288395\n",
            "2   52.5  169.3      68.0    68.7  ...  -0.000003  6.52818  84.908230 -16.208230\n",
            "3   31.6  138.5      62.9    69.1  ...  -0.000003  6.52818  86.726352 -17.626352\n",
            "4   30.7  140.5      59.2    69.7  ...  -0.000003  6.52818  84.955086 -15.255086\n",
            "5   29.4  141.3      55.2    69.9  ...  -0.000003  6.52818  83.503810 -13.603810\n",
            "6   36.5  150.5      59.8    69.9  ...  -0.000003  6.52818  83.688286 -13.788286\n",
            "7   31.4  134.0      59.4    70.0  ...  -0.000003  6.52818  88.575388 -18.575388\n",
            "8   29.2  137.5      56.4    70.1  ...  -0.000003  6.52818  85.314179 -15.214179\n",
            "9   35.2  144.8      64.7    70.5  ...  -0.000003  6.52818  85.986703 -15.486703\n",
            "\n",
            "[10 rows x 44 columns]\n",
            "   BMXWT  BMXHT  BMXWAIST  ...  sd_BMXHIP_7  mean_md_7   sd_md_7\n",
            "0   57.9  168.2      68.0  ...    13.955757   0.000041  5.355683\n",
            "1   34.0  145.4      60.6  ...    13.955757   0.000041  5.355683\n",
            "2   52.5  169.3      68.0  ...    13.955757   0.000041  5.355683\n",
            "3   31.6  138.5      62.9  ...    13.955757   0.000041  5.355683\n",
            "4   30.7  140.5      59.2  ...    13.955757   0.000041  5.355683\n",
            "5   29.4  141.3      55.2  ...    13.955757   0.000041  5.355683\n",
            "6   36.5  150.5      59.8  ...    13.955757   0.000041  5.355683\n",
            "7   31.4  134.0      59.4  ...    13.955757   0.000041  5.355683\n",
            "8   29.2  137.5      56.4  ...    13.955757   0.000041  5.355683\n",
            "9   35.2  144.8      64.7  ...    13.955757   0.000041  5.355683\n",
            "\n",
            "[10 rows x 48 columns]\n",
            "Mean md_7: 4.144073241301843e-05\n",
            "SD md_7:  5.355683342374397\n",
            "LLA:  -10.497097910321404\n",
            "ULA:  10.497180791786231\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "LOAA_7: 20.994278702107636\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IeMf-X6se5wu",
        "outputId": "2afdb6ea-3e32-4608-9fce-930b2091417a"
      },
      "source": [
        "#CC between models and BMXHIP:\n",
        "print(pearsonr(df[\"BMXHIP\"], df[\"BMXHIP_1\"]))\n",
        "print(pearsonr(df[\"BMXHIP\"], df[\"BMXHIP_2\"]))\n",
        "print(pearsonr(df[\"BMXHIP\"], df[\"BMXHIP_3\"]))\n",
        "print(pearsonr(df[\"BMXHIP\"], df[\"BMXHIP_4\"]))\n",
        "print(pearsonr(df[\"BMXHIP\"], df[\"BMXHIP_5\"]))\n",
        "print(pearsonr(df[\"BMXHIP\"], df[\"BMXHIP_6\"]))\n",
        "print(pearsonr(df[\"BMXHIP\"], df[\"BMXHIP_7\"]))"
      ],
      "execution_count": 54,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(0.10838995657869945, 5.2000082719719126e-27)\n",
            "(0.8789194297812091, 0.0)\n",
            "(0.8762186137937759, 0.0)\n",
            "(0.933173991318698, 0.0)\n",
            "(0.8785168216818459, 0.0)\n",
            "(0.899596342497604, 0.0)\n",
            "(0.9336124033231985, 0.0)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CKE-i_IBe7KK",
        "outputId": "7327cc17-f418-48af-ae04-d3ff5466d9e7"
      },
      "source": [
        "#Z-Score:\n",
        "df[\"mean_BMXHIP\"] = statistics.mean(df[\"BMXHIP\"])\n",
        "print(\"Mean BMXHIP: \", df[\"mean_BMXHIP\"])\n",
        "\n",
        "df[\"sd_BMXHIP\"] = statistics.stdev(df[\"BMXHIP\"])\n",
        "print(\"SD BMXHIP: \", df[\"sd_BMXHIP\"])\n",
        "\n",
        "df[\"zscore_BMXHIP\"] = ((df[\"BMXHIP\"])-(df[\"mean_BMXHIP\"]))/(df[\"sd_BMXHIP\"])\n",
        "print(\"Z-score BMXHIP: \", df[\"zscore_BMXHIP\"])\n",
        "\n",
        "df[\"mean_zscore_BMXHIP\"] = statistics.mean(df[\"zscore_BMXHIP\"])\n",
        "print(\"Mean z-score BMXHIP: \", df[\"mean_zscore_BMXHIP\"])"
      ],
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mean BMXHIP:  0       105.650332\n",
            "1       105.650332\n",
            "2       105.650332\n",
            "3       105.650332\n",
            "4       105.650332\n",
            "           ...    \n",
            "9798    105.650332\n",
            "9799    105.650332\n",
            "9800    105.650332\n",
            "9801    105.650332\n",
            "9802    105.650332\n",
            "Name: mean_BMXHIP, Length: 9803, dtype: float64\n",
            "SD BMXHIP:  0       14.948124\n",
            "1       14.948124\n",
            "2       14.948124\n",
            "3       14.948124\n",
            "4       14.948124\n",
            "          ...    \n",
            "9798    14.948124\n",
            "9799    14.948124\n",
            "9800    14.948124\n",
            "9801    14.948124\n",
            "9802    14.948124\n",
            "Name: sd_BMXHIP, Length: 9803, dtype: float64\n",
            "Z-score BMXHIP:  0      -2.612390\n",
            "1      -2.558872\n",
            "2      -2.471904\n",
            "3      -2.445145\n",
            "4      -2.405006\n",
            "          ...   \n",
            "9798    4.806601\n",
            "9799    4.893568\n",
            "9800    4.906948\n",
            "9801    4.973846\n",
            "9802    5.475581\n",
            "Name: zscore_BMXHIP, Length: 9803, dtype: float64\n",
            "Mean z-score BMXHIP:  0       6.213555e-17\n",
            "1       6.213555e-17\n",
            "2       6.213555e-17\n",
            "3       6.213555e-17\n",
            "4       6.213555e-17\n",
            "            ...     \n",
            "9798    6.213555e-17\n",
            "9799    6.213555e-17\n",
            "9800    6.213555e-17\n",
            "9801    6.213555e-17\n",
            "9802    6.213555e-17\n",
            "Name: mean_zscore_BMXHIP, Length: 9803, dtype: float64\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 483
        },
        "id": "dzScuU3GfMUI",
        "outputId": "ae22e248-550a-4845-b83d-b3912b277f5a"
      },
      "source": [
        "#Model 1:\n",
        "df[\"zscore_BMXHIP_1\"] = ((df[\"BMXHIP_1\"])-(df[\"mean_BMXHIP\"]))/(df[\"sd_BMXHIP\"])\n",
        "df[\"mean_zscore_BMXHIP_1\"] = statistics.mean(df[\"zscore_BMXHIP_1\"])\n",
        "print(\"Mean Z-score BMXHIP 1: \", df[\"mean_zscore_BMXHIP_1\"])\n",
        "\n",
        "#Graph:\n",
        "sns.kdeplot(df[\"zscore_BMXHIP\"], shade=True, color=\"orange\")\n",
        "kde = gaussian_kde(df[\"zscore_BMXHIP_1\"])\n",
        "x_range = linspace(min(df[\"zscore_BMXHIP_1\"]), max(df[\"zscore_BMXHIP_1\"]), len(df[\"zscore_BMXHIP_1\"]))\n",
        "sns.lineplot(x=x_range*-1, y=kde(x_range) * -1, color='lightgreen') \n",
        "plt.fill_between(x_range*-1, kde(x_range) * -1, color='lightgreen')\n",
        "plt.xlabel(\"Z-Score\")\n",
        "plt.axhline(y=0, linestyle='-',linewidth=1, color='black')\n",
        "plt.show()"
      ],
      "execution_count": 56,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mean Z-score BMXHIP 1:  0       0.000029\n",
            "1       0.000029\n",
            "2       0.000029\n",
            "3       0.000029\n",
            "4       0.000029\n",
            "          ...   \n",
            "9798    0.000029\n",
            "9799    0.000029\n",
            "9800    0.000029\n",
            "9801    0.000029\n",
            "9802    0.000029\n",
            "Name: mean_zscore_BMXHIP_1, Length: 9803, dtype: float64\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYoAAAEGCAYAAAB7DNKzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3de5hcdZ3n8fen+pr7HciVREQQuWoTZXEdFUREJCqygHJ1MDIrozvOsyyIq6gzu16edfRZWSWiPshkFmdwENQgEAWdHUHSaLiEOxFJQiAdAuESkpDOd/84p03RVFdXd9epc7rr83qeerrOqdOnfql01ad+l/P7KSIwMzMbSCnvApiZWbE5KMzMrCoHhZmZVeWgMDOzqhwUZmZWVWveBcjCzJkzY+HChXkXw8xs1Ljzzjs3R8SsSo+NyaBYuHAh3d3deRfDzGzUkPSngR5z05OZmVXloDAzs6ocFGZmVpWDwszMqso1KCQdL+lBSY9IuqjC4+dI6pG0Or2dl0c5zcyaWW6jniS1AJcB7wLWA6skXR8R9/U79EcRcUHDC2hmZkC+NYrFwCMRsTYidgJXA0tyLI+ZmVWQZ1DMBdaVba9P9/V3sqS7JV0jaf5AJ5O0VFK3pO6enp56l9XMrGkV/YK7nwL/NyJ2SPo4cCXwzkoHRsQyYBlAV1eXF9koggjYeh9svg22roEX/wg7n4XdO4EStI6D1snQPhU694LOvWHcbBi/ACbtBx2zQMr7X2HW9PIMig1AeQ1hXrrvzyLi6bLNK4CvNqBcNlLPPQyPXA6P/ygJi6mHwIQFMPUwaJ0EpdZk/+4dsGsb7HoetvfA8w/Dzmdg+ybYtg7UAtPfBHv9Bcw5AaYd7uAwy0GeQbEK2F/SIpKAOA34cPkBkmZHxMZ08yTg/sYW0Ybk6VVwzxeSGsTsd8MhX4AJi4b34R4BO7fAc/fDM6uT4FEJFp0F+30MJgzYCmlmdZZbUETELkkXADcCLcD3I2KNpC8C3RFxPfBJSScBu4AtwDl5ldeqeO5hWH1hEhALToXX/TW0dIzsnBJ0zIBZb01u8Z/h+Ydg402w4hDY+51w0H+DmW+uz7/BzAaksbhmdldXV3hSwAZ4+bmkBvHoD2DBh2D+B6GlM/vn3fUSbLwB1v0YJu0Ph1wKe789++c1G8Mk3RkRXZUe85XZNnQR8Mfl8NMD4LkH4c1XwMIPNyYkIOkEn/9BeMuVMOMtcNvZcNNb4albkrKZWV0VfdSTFc2za2DV+bBjMxz8WZjyhvzKUmqFOcfDPu+Cp1bCbefC+Hlw6Bdh73e449usThwUVptdL8Ldl8La78HCs2De+5JRSUVQakk6z/c+Fp76Jdz+0WSY7SGXwuzjHBhmI+SgsMFtWJHUIia/HhZfAR3T8y5RZaWWJBj2PgY23QLdn4DWifCGi2H+yUkNxMyGzO8cG9j2zdB9AWz+d3jdp2BGxX6u4im1wD7HJiOjNt8Ga/4n/OFCOOCTsN9fJhf4mVnN3Jltla2/Dn7+BiBg8XdHT0iUUwlmHQ1v+kYylHbjjXDdvnD7efB0tzu+zWrkGoW90ssvQPcn4cmbks7qqYfkXaL6mHJQctuxJRla+5v3Q/u0pIax8CPQWXFNeTPDNQort+VOuOFw2PEULF42dkKiXMf0JBiO+iG85lx44ga4fj+49UR4/Bro3Z53Cc0KxzUKg9gND/wDrPkf8LpPJG37Y51KMP2NyW3XNtj0G7jvK/C7jyXXaCw6A2a9LenvMGtyDopm99KTcNtZyc+ubyXDSptN6/jkeow5xycTEj51C9zxV/Dys0loLDgFZv1Hj5qypuW//Ga2/jr43dLkA/L1F/qDEJLpzvc9Nbm9+Dj0/BusugBeeiJ5neYuSYbgFnWIsFkGPNdTM9qxBe78FGy6NQmIsdgXUW/be5KhtltWJbPZTn59MvX57ONgxmKHrI161eZ68l93M4mAx5bD7/82mZH1yGXJvEk2uM5ZMO+k5Na7E7beA1u64U9XJ7WNmW9J1s2YeRRMPyIZUWU2RjgomkXPb+H3n05mfD3k88lQURuelvZkQaXpb0q2d26FZ++CZ+9OZrR97iFom5TMbDtxP5iwbzIH1bg5MH4OjJ8P7dM9tYiNGg6Ksa7n3+HeL8EzdyWL/sx+dzLix+qnfQrs9bbkBskosu2bYNt62L4RXlib1D52bkmasLY/mRw3YRFMPgCmHgxTD01uExf5/8cKx0ExFu3cCuuugYcugx09MP8UOODTyTdhy55KMG6f5DaQl5+HlzbAi+uSGsiTK+H5R5MaX194TDnolbUSTz1iOck1KCQdD3yTZIW7KyLiy/0e7wB+CLwJeBo4NSIea3Q5Cy8CXvxj8mGz/jrY9G/J9QELTkk6Wosyy6vt0TYJ2g6EyQe+cv/LLyT/ly8+Blt+n0zI+NIT8NLGpKlq3Nw9zVfj5ydrkffdHz8/qd2Y1VluQSGpBbgMeBewHlgl6fqIuK/ssL8EnomI10o6DfgKcGrjS1sgu3uTD5Gt9yXt4k+vgqfvgOiFaUfAjCPhtecnH0Q2+rRNTEah9R+JFpFM9b5jU7IWyPaepGnrmbuTWuOOTcm1MColfSHj5sD4udA5Oxny2zEz6WBvnwptk6F1UvI30joxuY7EzV1WRZ41isXAIxGxFkDS1cASoDwolgCXpvevAb4lSTFWxvRGQO9LyZXBvS8m3yZ3PZ80He18Jv1AeBK2bYBt65KA2LY+GcM/YVHSHDH10GTMf+fsUdc5GgQiuzLvJtjKTsbTSgejvFYlJSHSNhEmvqbyMRGw64U0ODYnw6B3boFtjydNXbteSMKmd1vyc1f6c/dOaBmXBEbrhFfeWsYnj7WMS1YwbOmEUkfSjKk2KLUlQ4PVAmpNbqXWZL/aoNSeHFvq3PP7LR3JOfrOUyo/V7rtK+ILJc+gmAusK9teD7x5oGMiYpekrcAMYHP/k0laCiwt2653eQtkY3r7bd4FsTHjxfTWk3dBrIDGTGd2RCwDlsEILrjbtS2Z62fTr/fse0XgKLlJZfdbkm8/fd+o+r5JtbSXfWvqTL+x9X0zG79nu0nnZXwZ+Om0Eus6sg/0Y57t5WDP9Te6xC7Y/XIySePu7dC7I72/I6kB7d6ZPB69e27Qb+r42PMzIt0uux+9Zfd3Vzi+TKUvnv1/N3YDu5P90Vu2v3fPuas1hpQ/R8V/xwDb5cfOPg6O/M6wBq5U+3KdZ1BsAOaXbc9L91U6Zr2kVmAKSad2NlrHw9HLMzu97dEGvD92c9VzV/Hs7mfrfn4hzpx8JtNapoGvfTMbkTy/zq4C9pe0SFI7cBpwfb9jrgfOTu9/CPjVmOmfMEoqcdS4ozI597yWeUlImNmI5VajSPscLgBuJBke+/2IWCPpi0B3RFwPfA+4StIjwBaSMLExZGHbwj93aMerqtjDd1jHYXU7l1mzy7WPIiJWACv67ftc2f3twCmNLpc1TrvamVGawebdrxqfMCLz2ufV9Xxmzaw5e1KtUGa31ncNjDba6FBHXc9p1swcFJa76S31XdthYmliXc9n1uwcFJa7egdFvc9n1uwcFJa7KaVkfqJ6XaU9ozSjLucxs4SDwnI3oTQBqGNQtDgozOrJQWG5a1Ur4zSubsNjp7R4BlWzenJQWCFMKk2qW1D01VDMrD4cFFYIk0uT63aucfI64Gb15KCwQqjXkFYhWrxQk1ldOSisEMaXxtflPK1jZ0Jks8JwUFghTFB9+hXa1FaX85jZHg4KK4R61Sg61VmX85jZHg4KK4Txqk9Q1CtwzGwPB4UVQt8HfGmEf5KTNKkexTGzMg4KK4S+Ia0jvTrbEwKa1Z+DwgqhRS20q33EF91NanGNwqzeHBRWGOM1fuRB4aYns7rLJSgkTZd0s6SH058VFzeW1CtpdXrrv562jTH1CAp3ZpvVX141iouAX0bE/sAv0+1KXoqIw9PbSY0rnuVhXGnkU294eKxZ/eUVFEuAK9P7VwLvz6kcViD1WL60o+QlUM3qLa+g2DsiNqb3nwT2HuC4Tkndkm6XVDVMJC1Nj+3u6empa2GtMeoRFO2016EkZlYus4lxJK0E9qnw0CXlGxERkgZqmN43IjZIeg3wK0n3RMSjlQ6MiGXAMoCurq76zFdtDTXSZiMhpPosfmRme2QWFBFx7ECPSXpK0uyI2ChpNrBpgHNsSH+ulXQrcARQMShs9BtpjWKkF+uZWWV5vbOuB85O758NXNf/AEnTpOSTQ9JM4GjgvoaV0BpupP0LLXh6cbMs5BUUXwbeJelh4Nh0G0ldkq5Ij3k90C3pLuAW4MsR4aAYw0ba9NQqTzFuloVc3lkR8TRwTIX93cB56f3fAoc0uGiWo5EGRbvckW2WBTfqWmH09VEMd74nB4VZNhwUVhh9H/TDDQpfbGeWDQeFFcZIaxQOCrNsOCisMFpoQWjY8z11lhwUZllwUFhhSKKN4a953YGn7zDLgoPCCmUka1K4M9ssGw4KK5SRBEU95ooys1dzUFihjOTD3jPHmmXDQWGFMpIPe/dRmGXDQWGF0k77sIfHuunJLBsOCiuUkXRIt5fcmW2WBQeFFUqHOobdmd2m4Q+tNbOBOSisUEZSo2jNZ45LszHPQWGFMpKgcI3CLBsOCiuUkQSFFy4yy4aDwgplJEHh9bLNsuGgsELxNBxmxZNLUEg6RdIaSbsldVU57nhJD0p6RNJFjSyj5aMvKEpD/NMc7rUXZja4vGoU9wIfBH4z0AGSWoDLgPcABwGnSzqoMcWzvAx38SIHhVl28loz+34YtE15MfBIRKxNj70aWALcl3kBLTfDbXpyUJhlp8h9FHOBdWXb69N9FUlaKqlbUndPT0/mhbNsDDcohtpUZWa1y6xGIWklsE+Fhy6JiOvq/XwRsQxYBtDV1TW8S3std30LFw316mwHhVl2MguKiDh2hKfYAMwv256X7rMxrG+Vu156h/x7ZpaNIn8NWwXsL2mRpHbgNOD6nMtkDdCmtiHXKHyxnVl28hoe+wFJ64GjgJ9LujHdP0fSCoCI2AVcANwI3A/8c0SsyaO81ljD6adw05NZdvIa9XQtcG2F/U8AJ5RtrwBWNLBoVgCttA69RiHXKMyy4q9hVjitGvr3Fzc9mWWnpqCQ9K+S3ivJwWKZG84ssMMJFzOrTa0f/P8H+DDwsKQvSzogwzJZkxtOM1LfsFozq7+agiIiVkbER4A3Ao8BKyX9VtK5khcBsPoazgJErlGYZafmpiRJM4BzgPOAPwDfJAmOmzMpmTUt91GYFUtN70hJ1wIHAFcB74uIjelDP5LUnVXhrDm10orQkEY+uUZhlp1a313fTYeq/pmkjojYEREDThNuNhytGvrwWAeFWXZqbXr6uwr7bqtnQcz6DKcZyetlm2Wn6tcwSfuQzNg6TtIR8Oe5nCcD4zMumzWp4dQOhtMBbma1Gezd9W6SDux5wNfL9j8PfCajMlmTG05QuEZhlp2q78iIuBK4UtLJEfHjBpXJmtxwmp5cozDLzmBNT2dExD8CCyV9uv/jEfH1Cr9mNiLDqlGUXKMwy8pg78gJ6c+JWRfErM9wagftDG9lPDMb3GBNT5enP7/QmOKYuY/CrGhqnRTwq5ImS2qT9EtJPZLOyLpw1pyG00cx3LW2zWxwtV5HcVxEPAecSDLX02uB/5pVoay5uUZhViy1BkXfO/e9wL9ExNaRPKmkUyStkbRb0oBXdkt6TNI9klZ7qpDmMZygcB+FWXZqfUf+TNIDwEvAX0maBWwfwfPeC3wQuLyGY98REZtH8Fw2ygynM9ujnsyyU9M7MiIukvRVYGtE9Ep6EVgy3CeNiPsBJA12qDWhvvUohjIxoGePNcvOUL66HUhyPUX57/ywzuXpL4CbJAVweUQsG+hASUuBpQALFizIuFiWpb4axZCCwmtmm2Wm1mnGrwL2A1YDvenuoEpQSFoJ7FPhoUsi4roay/fWiNggaS/gZkkPRMRvKh2YhsgygK6urqFNPWqF4vUozIql1ndkF3BQRNT8ARwRxw6vSK84x4b056Z0TYzFQMWgsLFjOH0UpdrX4DKzIar13XUvlWsHmZE0QdKkvvvAcWk5bIwr76Oolfu7zLJT61e3mcB9ku4AdvTtjIiThvOkkj4A/G9gFvBzSasj4t2S5gBXRMQJwN7AtekHQCvwTxHxi+E8n40ufc1IQ128yMyyUWtQXFrPJ42Ia4FrK+x/Ajghvb8WOKyez2ujgyRaaKk5KIZS8zCzoat1eOyvJe0L7B8RKyWNB/ceWnZaaGEXu/IuhplR+1xPHwOuYc8FcnOBn2RVKLOhDHd1jcIsW7V2Zn8COBp4DiAiHgb2yqpQZq20uunJrCBqDYodEbGzbyO96M49jZYZ1yjMiqPWoPi1pM8A4yS9C/gX4KfZFcuaXZvaaq5R+BoKs2zV+g67COgB7gE+DqwAPptVocyGctGdg8IsW7WOetot6SfATyKiJ+MymQ1pGo+SHBRmWar6DlPiUkmbgQeBB9PV7T7XmOJZs2qj9mnDXaMwy9Zg77C/IRntdGRETI+I6cCbgaMl/U3mpbOmNaQahYPCLFODvcPOBE6PiD/27UivmD4DOCvLgllzG8qoJzc9mWVrsHdYW6XV5dJ+Ci8pZpkZSo1iOLPNmlntBguKncN8zGxEhvLh70WLzLI12LvxMEnPVdgvoDOD8pgBrlGYFUnVd1hE+Kua5WIoH/7DWRHPzGrnXkArpKE0J7lGYZYtB4UVkmsUZsXhoLBCGlIfhYPCLFO5BIWkr0l6QNLdkq6VNHWA446X9KCkRyRd1OhyWn7cmW1WHHnVKG4GDo6IQ4GHgIv7HyCpBbgMeA9wEHC6pIMaWkrLTcsQFlBsK/mSHrMs5RIUEXFTRPStc3k7MK/CYYuBRyJibboWxtXAkkaV0fI1lBrFUELFzIauCH0UHwVuqLB/LrCubHt9uq8iSUsldUvq7unxBLej3VCak9ppz7AkZpZZ466klcA+FR66JCKuS4+5BNgFLB/p80XEMmAZQFdXl1ffG+WGUqNok5uezLKUWVBExLHVHpd0DnAicExEVPpg3wDML9uel+6zJjCkPgoHhVmm8hr1dDxwIXBSRGwb4LBVwP6SFklqB04Drm9UGS1ffTWKWtbDdlCYZSuvPopvAZOAmyWtlvQdAElzJK0ASDu7LwBuBO4H/jki1uRUXmuwIQWFJzI2y1QuA9Aj4rUD7H8COKFsewXJ+tzWZIbSme3hsWbZKsKoJ7NX6WtOqqVG4eGxZtlyUFgh9X34B4MPYPMKd2bZ8jvMCklSzc1PrlGYZctBYYVV67UUXuHOLFsOCiusNtpqanrypIBm2XJQWGG5RmFWDA4KK6x2tddUo3AfhVm2HBRWWLVece2gMMuWg8IKq9ZZYaXBr7Uws+FzUFhh+Yprs2JwUFhh1TKHUy1XbpvZyDgorLBq6aNwUJhlz0FhheWgMCsGB4UVVi3XUZT8J2yWOb/LrLBq6aNwUJhlz+8yK6xamp48c6xZ9vwus8KqKSj8J2yWuVxmU5P0NeB9wE7gUeDciHi2wnGPAc8DvcCuiOhqZDktX7VM9uerss2yl9fXsZuBgyPiUOAh4OIqx74jIg53SDSfdg1+ZbYnBDTLXi5BERE3RcSudPN2YF4e5bBiq2XUk6cYN8teERp4PwrcMMBjAdwk6U5JS6udRNJSSd2Sunt6eupeSGu8WtbNrnUqcjMbvszeZZJWAvtUeOiSiLguPeYSYBewfIDTvDUiNkjaC7hZ0gMR8ZtKB0bEMmAZQFdX1+BzU1vhlQfFQNONOyjMspfZuywijq32uKRzgBOBYyKi4qdARGxIf26SdC2wGKgYFDb2dKhj0GNq6ccws5HJpelJ0vHAhcBJEbFtgGMmSJrUdx84Dri3caW0vNUyzXgtF+WZ2cjk1UfxLWASSXPSaknfAZA0R9KK9Ji9gf8n6S7gDuDnEfGLfIpreZA0aGe1axRm2culgTciXjvA/ieAE9L7a4HDGlkuK552tbM9tg/4eK2r4JnZ8BVh1JPZgAZbN9vDY82y56CwQutUZ9WgaC+56cksaw4KK7TBRj65j8Isew4KK7TBgqKTzgaVxKx5OSis0AYLCl9wZ5Y9B4UV2mBNS+6jMMueg8IKbdAahUc9mWXOQWGFNliNwk1PZtlzUFihuY/CLH8OCiu0wWoUnuvJLHsOCiu0wWoUnsLDLHsOCiu0zlJyncRAixd5zWyz7DkorNAmaAJQOSiEkAZe/c7M6sNBYYXWoQ5KA/yZVlsi1czqx0FhhSaJcRpX+TEHhVlDOCis8CaUJrCb3a/a7/4Js8bILSgkfUnS3ekKdzdJmjPAcWdLeji9nd3oclr+ppSmVNzfIgeFWSPkWaP4WkQcGhGHAz8DPtf/AEnTgc8DbwYWA5+XNK2xxbS8TS5Nrrjf03eYNUZuQRERz5VtToCKq9O8G7g5IrZExDPAzcDxjSifFcfklgGCwldlmzVEru80SX8PnAVsBd5R4ZC5wLqy7fXpvkrnWgosBViwYEF9C2q5GqhG4UWLzBoj0xqFpJWS7q1wWwIQEZdExHxgOXDBSJ4rIpZFRFdEdM2aNasexbeC6AuK/p3Xg121bWb1kWmNIiKOrfHQ5cAKkv6IchuAt5dtzwNuHXHBbFQZqEYx0LBZM6uvPEc97V+2uQR4oMJhNwLHSZqWdmIfl+6zJtKqVjrV+aohsuNL43MqkVlzybOP4suSDgB2A38CzgeQ1AWcHxHnRcQWSV8CVqW/88WI2JJPcS1Pk0qT2N67/RX7OuX1ss0aIbegiIiTB9jfDZxXtv194PuNKpcV0wRNoIeeV+xzjcKsMXxlto0KlUJhvBwUZo3goLBRoVJQ9E1BbmbZclDYqFCp9uA+CrPGcFDYqFCpRuHrKMwaw0Fho0KlayZcozBrDAeFjQr9axRCnj3WrEEcFDYq9O+j8FoUZo3joLBRoa+ZqW9VO08IaNY4DgobFUoq0c6ecPA1FGaN46CwUaOz1Emky5ZMbZmac2nMmoeDwkaN8lrEXi175VgSs+bioLBRY2Jp4p/vz2rxmiNmjeKgsFGjPCimt07PsSRmzcVBYaPG1NKefolJmpRjScyai1ent1HjwI4DeWDnA0wpTUFS3sUxaxoOChs1OtTBqZNPzbsYZk3HTU9mZlZVLjWKdHnTJSTLoG4CzomIJyoc1wvck24+HhEnNa6UZmYG+dUovhYRh0bE4cDPgM8NcNxLEXF4enNImJnlIJegiIjnyjYnQHq5rZmZFU5ufRSS/l7SOuAjDFyj6JTULel2Se8f5HxL02O7e3p66l5eM7NmpYhsvsxLWgnsU+GhSyLiurLjLgY6I+LzFc4xNyI2SHoN8CvgmIh4dLDn7urqiu7u7hGU3sysuUi6MyK6Kj2WWWd2RBxb46HLgRXAq4IiIjakP9dKuhU4Ahg0KMzMrH5yaXqStH/Z5hLggQrHTJOSRZElzQSOBu5rTAnNzKxPZk1PVZ9U+jFwAMnw2D8B56dNTF3p/fMk/Qfg8vSYEvCNiPhejefvSc87FswENuddiALx67GHX4s9/Fq80nBej30jouJsm7kEhdVOUvdA7YbNyK/HHn4t9vBr8Ur1fj18ZbaZmVXloDAzs6ocFMW3LO8CFIxfjz38Wuzh1+KV6vp6uI/CzMyqco3CzMyqclCYmVlVDopRQNLXJD0g6W5J10qaOvhvjS2Sjpf0oKRHJF2Ud3nyJGm+pFsk3SdpjaRP5V2mvElqkfQHST/Luyx5kjRV0jXp58X9ko6qx3kdFKPDzcDBEXEo8BBwcc7laShJLcBlwHuAg4DTJR2Ub6lytQv424g4CHgL8Ikmfz0APgXcn3chCuCbwC8i4kDgMOr0mjgoRoGIuCkidqWbtwPz8ixPDhYDj0TE2ojYCVxNMvVLU4qIjRHx+/T+8yQfBnPzLVV+JM0D3gtckXdZ8iRpCvA24HsAEbEzIp6tx7kdFKPPR4Eb8i5Eg80F1pVtr6eJPxjLSVpIMlnm7/ItSa6+AVxIMt1PM1sE9AA/SJvhrpA0oR4ndlAUhKSVku6tcFtSdswlJM0Oy/MrqRWFpInAj4H/0m8xsKYh6URgU0TcmXdZCqAVeCPw7Yg4AngRqEt/Xi5rZturDTYtu6RzgBNJ1uRototfNgDzy7bnpfualqQ2kpBYHhH/mnd5cnQ0cJKkE4BOYLKkf4yIM3IuVx7WA+sjoq92eQ11CgrXKEYBSceTVK1PiohteZcnB6uA/SUtktQOnAZcn3OZciNJJO3Q90fE1/MuT54i4uKImBcRC0n+Ln7VpCFBRDwJrJN0QLrrGOq0NINrFKPDt4AO4ObkM4LbI+L8fIvUOBGxS9IFwI1AC/D9iFiTc7HydDRwJnCPpNXpvs9ExIocy2TF8NfA8vQL1Vrg3Hqc1FN4mJlZVW56MjOzqhwUZmZWlYPCzMyqclCYmVlVDgozM6vKQWE2CEkfkLS63223pPf0O+4tkn6XPn6/pEtzKrJZXXl4rNkQSVoKfAR4R0TsLtv/IPCfIuKudMbbAyJiRBc8SWqJiN6RldhsZFyjMBsCSa8DPgecWR4Sqb2AjQAR0dsXEpImSvqBpHvSNUVOTvefnu67V9JXyp7jBUn/S9JdwFGSzpB0R1pTuTwNIbOGcVCY1SidX+mfSNaCeLzCIf8APJguLvVxSZ3p/v8ObI2IQ9I1RX4laQ7wFeCdwOHAkZLenx4/AfhdRBwGPA2cChwdEYcDvSS1GbOGcVCY1e5LwJqI+FGlByPii0AXcBPwYeAX6UPHkiy81HfcM8CRwK0R0ZOuNbKcZC0BSMLgx+n9Y4A3AavS6TqOAV5Tz3+U2WA815NZDSS9HTiZZBrnvn0/IFkL4omIOAEgIh4Fvi3pu0CPpBnDeLrtZf0SAq6MiKZa1dCKxTUKs0FImgb8ADgrXVEOgIg4NyIO7wsJSe9NZ3YF2J+kZvAsyVK2n+h3vjuAv5A0M+1zOB34dYWn/yXwIUl7pb87XdK+df9HmlXhoDAb3PkkHdXf7jdE9tR+x51J0kexGrgK+EhaM/g7YFraaX0XyWipjSRrBdwC3AXcGRHX9X/itAB6lPYAAABESURBVEP8s8BNku4mCZ3ZGf07zSry8FgzM6vKNQozM6vKQWFmZlU5KMzMrCoHhZmZVeWgMDOzqhwUZmZWlYPCzMyq+v/NUU49DXdHoAAAAABJRU5ErkJggg==\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 483
        },
        "id": "AqUxW1C3fqha",
        "outputId": "b42f109d-bc27-4d77-d136-e54391bb9062"
      },
      "source": [
        "#Model 2:\n",
        "df[\"zscore_BMXHIP_2\"] = ((df[\"BMXHIP_2\"])-(df[\"mean_BMXHIP\"]))/(df[\"sd_BMXHIP\"])\n",
        "df[\"mean_zscore_BMXHIP_2\"] = statistics.mean(df[\"zscore_BMXHIP_2\"])\n",
        "print(\"Mean Z-score BMXHIP 2: \", df[\"mean_zscore_BMXHIP_2\"])\n",
        "#\n",
        "#Graph:\n",
        "sns.kdeplot(df[\"zscore_BMXHIP\"], shade=True, color=\"orange\")\n",
        "kde = gaussian_kde(df[\"zscore_BMXHIP_2\"])\n",
        "x_range = linspace(min(df[\"zscore_BMXHIP_2\"]), max(df[\"zscore_BMXHIP_2\"]), len(df[\"zscore_BMXHIP_2\"]))\n",
        "sns.lineplot(x=x_range*-1, y=kde(x_range) * -1, color='lightgreen') \n",
        "plt.fill_between(x_range*-1, kde(x_range) * -1, color='lightgreen')\n",
        "plt.xlabel(\"Z-Score\")\n",
        "plt.axhline(y=0, linestyle='-',linewidth=1, color='black')\n",
        "plt.show()"
      ],
      "execution_count": 57,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mean Z-score BMXHIP 2:  0       4.557228e-09\n",
            "1       4.557228e-09\n",
            "2       4.557228e-09\n",
            "3       4.557228e-09\n",
            "4       4.557228e-09\n",
            "            ...     \n",
            "9798    4.557228e-09\n",
            "9799    4.557228e-09\n",
            "9800    4.557228e-09\n",
            "9801    4.557228e-09\n",
            "9802    4.557228e-09\n",
            "Name: mean_zscore_BMXHIP_2, Length: 9803, dtype: float64\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    }
  ]
}