{
"metadata": {
"name": "model_checking_and_selection"
},
"nbformat": 3,
"nbformat_minor": 0,
"worksheets": [
{
"cells": [
{
"cell_type": "code",
"collapsed": false,
"input": [
"import pandas as pd\n",
"import numpy as np"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 4
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"%load_ext rmagic"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 3
},
{
"cell_type": "heading",
"level": 4,
"metadata": {},
"source": [
"Robust ..."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"%%R -o data\n",
"set.seed(3433); par(mfrow=c(1,2)); \n",
"data <- rnorm(100,mean=seq(0,3,length=100),sd=seq(0.1,3,length=100))"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 5
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"df = pd.DataFrame(zip(data, np.linspace(0, 3, num=100)), columns=['data', 'x'])"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 33
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"from statsmodels.formula.api import ols\n",
"from statsmodels.stats.sandwich_covariance import cov_hc3\n",
"\n",
"lm1 = ols('data ~ x', df).fit()\n",
"\n",
"cov_hc3(lm1)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "pyout",
"prompt_number": 31,
"text": [
"array([[ 0.05410024, -0.0476481 ],\n",
" [-0.0476481 , 0.05368603]])"
]
}
],
"prompt_number": 31
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"lm1.normalized_cov_params"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"\n",
"
\n",
" \n",
" \n",
" | \n",
" Intercept | \n",
" x | \n",
"
\n",
" \n",
" \n",
" \n",
" | Intercept | \n",
" 0.039406 | \n",
" -0.019604 | \n",
"
\n",
" \n",
" | x | \n",
" -0.019604 | \n",
" 0.013069 | \n",
"
\n",
" \n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
" | \n",
" X | \n",
" score | \n",
" rating | \n",
" genre | \n",
" box_office | \n",
" running_time | \n",
"
\n",
" \n",
" \n",
" \n",
" | 0 | \n",
" 2 Fast 2 Furious | \n",
" 48.9 | \n",
" PG-13 | \n",
" action/adventure | \n",
" 127.146 | \n",
" 107 | \n",
"
\n",
" \n",
" | 1 | \n",
" 28 Days Later | \n",
" 78.2 | \n",
" R | \n",
" horror | \n",
" 45.065 | \n",
" 113 | \n",
"
\n",
" \n",
" | 2 | \n",
" A Guy Thing | \n",
" 39.5 | \n",
" PG-13 | \n",
" rom comedy | \n",
" 15.545 | \n",
" 101 | \n",
"
\n",
" \n",
" | 3 | \n",
" A Man Apart | \n",
" 42.9 | \n",
" R | \n",
" action/adventure | \n",
" 26.248 | \n",
" 110 | \n",
"
\n",
" \n",
" | 4 | \n",
" A Mighty Wind | \n",
" 79.9 | \n",
" PG-13 | \n",
" comedy | \n",
" 17.781 | \n",
" 91 | \n",
"
\n",
" \n",
"
\n",
"
\n",
"OLS Regression Results\n",
"\n",
" | Dep. Variable: | score | R-squared: | 0.212 | \n",
"
\n",
"\n",
" | Model: | OLS | Adj. R-squared: | 0.201 | \n",
"
\n",
"\n",
" | Method: | Least Squares | F-statistic: | 18.46 | \n",
"
\n",
"\n",
" | Date: | Thu, 21 Feb 2013 | Prob (F-statistic): | 7.98e-08 | \n",
"
\n",
"\n",
" | Time: | 23:49:35 | Log-Likelihood: | -554.61 | \n",
"
\n",
"\n",
" | No. Observations: | 140 | AIC: | 1115. | \n",
"
\n",
"\n",
" | Df Residuals: | 137 | BIC: | 1124. | \n",
"
\n",
"\n",
" | Df Model: | 2 | | | \n",
"
\n",
"
\n",
"\n",
"\n",
" | coef | std err | t | P>|t| | [95.0% Conf. Int.] | \n",
"
\n",
"\n",
" | Intercept | 37.2364 | 5.606 | 6.643 | 0.000 | 26.152 48.321 | \n",
"
\n",
"\n",
" | box_office | 0.0824 | 0.018 | 4.495 | 0.000 | 0.046 0.119 | \n",
"
\n",
"\n",
" | running_time | 0.1275 | 0.054 | 2.379 | 0.019 | 0.022 0.234 | \n",
"
\n",
"
\n",
"\n",
"\n",
" | Omnibus: | 2.420 | Durbin-Watson: | 2.124 | \n",
"
\n",
"\n",
" | Prob(Omnibus): | 0.298 | Jarque-Bera (JB): | 2.230 | \n",
"
\n",
"\n",
" | Skew: | 0.309 | Prob(JB): | 0.328 | \n",
"
\n",
"\n",
" | Kurtosis: | 2.991 | Cond. No. | 677. | \n",
"
\n",
"
"
],
"output_type": "pyout",
"prompt_number": 74,
"text": [
"\n",
"\"\"\"\n",
" OLS Regression Results \n",
"==============================================================================\n",
"Dep. Variable: score R-squared: 0.212\n",
"Model: OLS Adj. R-squared: 0.201\n",
"Method: Least Squares F-statistic: 18.46\n",
"Date: Thu, 21 Feb 2013 Prob (F-statistic): 7.98e-08\n",
"Time: 23:49:35 Log-Likelihood: -554.61\n",
"No. Observations: 140 AIC: 1115.\n",
"Df Residuals: 137 BIC: 1124.\n",
"Df Model: 2 \n",
"================================================================================\n",
" coef std err t P>|t| [95.0% Conf. Int.]\n",
"--------------------------------------------------------------------------------\n",
"Intercept 37.2364 5.606 6.643 0.000 26.152 48.321\n",
"box_office 0.0824 0.018 4.495 0.000 0.046 0.119\n",
"running_time 0.1275 0.054 2.379 0.019 0.022 0.234\n",
"==============================================================================\n",
"Omnibus: 2.420 Durbin-Watson: 2.124\n",
"Prob(Omnibus): 0.298 Jarque-Bera (JB): 2.230\n",
"Skew: 0.309 Prob(JB): 0.328\n",
"Kurtosis: 2.991 Cond. No. 677.\n",
"==============================================================================\n",
"\"\"\""
]
}
],
"prompt_number": 74
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# no regsubsets\n",
"# no bic.glm"
],
"language": "python",
"metadata": {},
"outputs": []
}
],
"metadata": {}
}
]
}