{ "cells": [ { "cell_type": "markdown", "metadata": { "collapsed": false }, "source": [ "# Statsmodels OLS fit" ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "collapsed": false }, "outputs": [ ], "source": [ "import pandas as pd" ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "collapsed": false }, "outputs": [ ], "source": [ "data = pd.DataFrame({\n", " 'x': [1, 2, 4, 5, 6, 7.7, 8.2, 9, 10, 11],\n", " 'y': [3, 3.6, 4.2, 4.1, 5, 4.9, 5.5, 6, 6.1, 6.5],\n", "})" ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "collapsed": false }, "outputs": [ ], "source": [ "import statsmodels.formula.api as smf" ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "collapsed": false }, "outputs": [ ], "source": [ "model = smf.ols('y ~ x', data=data)" ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "collapsed": false }, "outputs": [ ], "source": [ "res = model.fit()" ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/plain": [ "0 2.741911\n", "1 4.422732\n", "2 6.103553\n", "3 36.358325\n", "dtype: float64" ] }, "execution_count": 6, "metadata": { }, "output_type": "execute_result" } ], "source": [ "res.predict(pd.DataFrame({'x': [0, 5, 10, 100]}))" ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "collapsed": false }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/usr/local/lib/python3.6/dist-packages/scipy/stats/stats.py:1535: UserWarning: kurtosistest only valid for n>=20 ... continuing anyway, n=10\n", " \"anyway, n=%i\" % int(n))\n" ] }, { "data": { "text/html": [ "\n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "
OLS Regression Results
Dep. Variable: y R-squared: 0.962
Model: OLS Adj. R-squared: 0.957
Method: Least Squares F-statistic: 201.4
Date: Sun, 28 Jun 2020 Prob (F-statistic): 5.91e-07
Time: 09:20:50 Log-Likelihood: 1.2199
No. Observations: 10 AIC: 1.560
Df Residuals: 8 BIC: 2.165
Df Model: 1
Covariance Type: nonrobust
\n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "
coef std err t P>|t| [0.025 0.975]
Intercept 2.7419 0.169 16.201 0.000 2.352 3.132
x 0.3362 0.024 14.193 0.000 0.282 0.391
\n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "
Omnibus: 2.060 Durbin-Watson: 2.907
Prob(Omnibus): 0.357 Jarque-Bera (JB): 1.180
Skew: -0.805 Prob(JB): 0.554
Kurtosis: 2.512 Cond. No. 16.2


Warnings:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified." ], "text/plain": [ "\n", "\"\"\"\n", " OLS Regression Results \n", "==============================================================================\n", "Dep. Variable: y R-squared: 0.962\n", "Model: OLS Adj. R-squared: 0.957\n", "Method: Least Squares F-statistic: 201.4\n", "Date: Sun, 28 Jun 2020 Prob (F-statistic): 5.91e-07\n", "Time: 09:20:50 Log-Likelihood: 1.2199\n", "No. Observations: 10 AIC: 1.560\n", "Df Residuals: 8 BIC: 2.165\n", "Df Model: 1 \n", "Covariance Type: nonrobust \n", "==============================================================================\n", " coef std err t P>|t| [0.025 0.975]\n", "------------------------------------------------------------------------------\n", "Intercept 2.7419 0.169 16.201 0.000 2.352 3.132\n", "x 0.3362 0.024 14.193 0.000 0.282 0.391\n", "==============================================================================\n", "Omnibus: 2.060 Durbin-Watson: 2.907\n", "Prob(Omnibus): 0.357 Jarque-Bera (JB): 1.180\n", "Skew: -0.805 Prob(JB): 0.554\n", "Kurtosis: 2.512 Cond. No. 16.2\n", "==============================================================================\n", "\n", "Warnings:\n", "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n", "\"\"\"" ] }, "execution_count": 7, "metadata": { }, "output_type": "execute_result" } ], "source": [ "res.summary()" ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "collapsed": false }, "outputs": [ ], "source": [ "import numpy as np\n", "import matplotlib.pyplot as plt" ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "collapsed": false }, "outputs": [ { "data": { "text/plain": [ "[]" ] }, "execution_count": 9, "metadata": { }, "output_type": "execute_result" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "execution_count": 9, "metadata": { "image/png": { "height": 424, "width": 710 }, "needs_background": "light" }, "output_type": "execute_result" } ], "source": [ "fig = data.plot.scatter('x', 'y', s=40, grid=True)\n", "xx = np.linspace(-2, 15, 100)\n", "yy = res.predict(pd.DataFrame({'x': xx}))\n", "fig.plot(xx, yy, color='green')" ] }, { "cell_type": "code", "execution_count": 10, "metadata": { "collapsed": false }, "outputs": [ ], "source": [ "data2 = pd.DataFrame({\n", " 'x1': [1, 2, 4, 5, 6, 7.7, 8.2, 9, 10, 11],\n", " 'x2': [4, 5, 4, 5, 6, 7.7, 7, 7.9, 8, 8.1],\n", " 'y': [3, 3.6, 4.2, 4.1, 5, 4.9, 5.5, 6, 6.1, 6.5],\n", "})" ] }, { "cell_type": "code", "execution_count": 11, "metadata": { "collapsed": false }, "outputs": [ ], "source": [ "model2 = smf.ols('y ~ x1 + x2', data=data2)\n", "res2 = model2.fit()" ] }, { "cell_type": "code", "execution_count": 12, "metadata": { "collapsed": false }, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/usr/local/lib/python3.6/dist-packages/scipy/stats/stats.py:1535: UserWarning: kurtosistest only valid for n>=20 ... continuing anyway, n=10\n", " \"anyway, n=%i\" % int(n))\n" ] }, { "data": { "text/html": [ "\n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "
OLS Regression Results
Dep. Variable: y R-squared: 0.962
Model: OLS Adj. R-squared: 0.951
Method: Least Squares F-statistic: 88.75
Date: Sun, 28 Jun 2020 Prob (F-statistic): 1.06e-05
Time: 09:20:54 Log-Likelihood: 1.2542
No. Observations: 10 AIC: 3.492
Df Residuals: 7 BIC: 4.399
Df Model: 2
Covariance Type: nonrobust
\n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "
coef std err t P>|t| [0.025 0.975]
Intercept 2.8433 0.496 5.730 0.001 1.670 4.017
x1 0.3503 0.069 5.063 0.001 0.187 0.514
x2 -0.0306 0.139 -0.219 0.833 -0.360 0.299
\n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "\n", " \n", "\n", "
Omnibus: 1.553 Durbin-Watson: 2.877
Prob(Omnibus): 0.460 Jarque-Bera (JB): 0.992
Skew: -0.707 Prob(JB): 0.609
Kurtosis: 2.383 Cond. No. 61.4


Warnings:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified." ], "text/plain": [ "\n", "\"\"\"\n", " OLS Regression Results \n", "==============================================================================\n", "Dep. Variable: y R-squared: 0.962\n", "Model: OLS Adj. R-squared: 0.951\n", "Method: Least Squares F-statistic: 88.75\n", "Date: Sun, 28 Jun 2020 Prob (F-statistic): 1.06e-05\n", "Time: 09:20:54 Log-Likelihood: 1.2542\n", "No. Observations: 10 AIC: 3.492\n", "Df Residuals: 7 BIC: 4.399\n", "Df Model: 2 \n", "Covariance Type: nonrobust \n", "==============================================================================\n", " coef std err t P>|t| [0.025 0.975]\n", "------------------------------------------------------------------------------\n", "Intercept 2.8433 0.496 5.730 0.001 1.670 4.017\n", "x1 0.3503 0.069 5.063 0.001 0.187 0.514\n", "x2 -0.0306 0.139 -0.219 0.833 -0.360 0.299\n", "==============================================================================\n", "Omnibus: 1.553 Durbin-Watson: 2.877\n", "Prob(Omnibus): 0.460 Jarque-Bera (JB): 0.992\n", "Skew: -0.707 Prob(JB): 0.609\n", "Kurtosis: 2.383 Cond. No. 61.4\n", "==============================================================================\n", "\n", "Warnings:\n", "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n", "\"\"\"" ] }, "execution_count": 12, "metadata": { }, "output_type": "execute_result" } ], "source": [ "res2.summary()" ] }, { "cell_type": "code", "execution_count": 0, "metadata": { "collapsed": false }, "outputs": [ ], "source": [ ] }, { "cell_type": "code", "execution_count": 0, "metadata": { "collapsed": false }, "outputs": [ ], "source": [ ] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (system-wide)", "language": "python", "metadata": { "cocalc": { "description": "Python 3 programming language", "priority": 100, "url": "https://www.python.org/" } }, "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.6.9" } }, "nbformat": 4, "nbformat_minor": 4 }