{ "cells": [ { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Populating the interactive namespace from numpy and matplotlib\n" ] } ], "source": [ "# import analysis library\n", "import pandas as pd, numpy as np \n", "import seaborn as sns\n", "import matplotlib.pyplot as plt\n", "%matplotlib inline\n", "%pylab inline\n", "plt.style.use('ggplot')" ] }, { "cell_type": "code", "execution_count": 120, "metadata": {}, "outputs": [], "source": [ "# create some fake data \n", "# sklearn requires the data shape of (row number, column number)\n", "# since shape(x) = (100,) originally, we reshape x to x_ here \n", "# https://stackoverflow.com/questions/30813044/sklearn-found-arrays-with-inconsistent-numbers-of-samples-when-calling-linearre\n", "\n", "\n", "x = np.linspace(0,100,50).reshape(50,1)\n", "a = 5\n", "b = 10\n", "#noise = np.random.rand(100).reshape(100,1)\n", "noise = np.random.randint(1000, size=50).reshape(50,1)\n", "# y = 0.1*x + noise \n", "y = a*x+noise" ] }, { "cell_type": "code", "execution_count": 158, "metadata": { "collapsed": true }, "outputs": [], "source": [ "def linear_model_(x,y,prob):\n", " from sklearn.linear_model import LinearRegression\n", " import scipy.stats\n", " \n", " regr = LinearRegression()\n", " regr.fit(x,y)\n", " plt.scatter(x,y, label='data')\n", " plt.plot(x,regr.predict(x),label='fitting line')\n", " plt.legend(loc='upper left')\n", " plt.xlabel('x')\n", " plt.xlabel('y')\n", " plt.title('Linear Regression : y=a*x+b')\n", " print ('-----------')\n", " print ('regression coefficient : ', regr.coef_)\n", " print ('regression interception : ', regr.intercept_)\n", " print ('residues : ', regr.residues_)\n", " # confidence interval\n", " # https://gist.github.com/riccardoscalco/5356167\n", " n = len(x)\n", " xy = x * y\n", " xx = x * x\n", " # \n", " b1 = (xy.mean() - x.mean() * y.mean()) / (xx.mean() - x.mean()**2)\n", " b0 = y.mean() - b1 * x.mean()\n", " s2 = 1./n * sum([(y[i] - b0 - b1 * x[i])**2 for i in range(n)])\n", " #print ('regression interception = ',b0)\n", " #print ('regression coefficient = ',b1)\n", " print ('s2 = ',s2)\n", " print ('')\n", " #prob = .95 \n", " alpha = 1 - prob\n", " c1 = scipy.stats.chi2.ppf(alpha/2.,n-2)\n", " c2 = scipy.stats.chi2.ppf(1-alpha/2.,n-2)\n", " print ('confidence interval of s2: ',[n*s2/c2,n*s2/c1])\n", " c = -1 * scipy.stats.t.ppf(alpha/2.,n-2)\n", " bb1 = c * (s2 / ((n-2) * (xx.mean() - (x.mean())**2)))**.5\n", " print ('confidence interval of regression coefficient: ',[b1-bb1,b1+bb1])\n", " \n", " bb0 = c * ((s2 / (n-2)) * (1 + (x.mean())**2 / (xx.mean() - (x.mean())**2)))**.5\n", " print ('confidence interval of regression interception: ',[b0-bb0,b0+bb0])\n", " print ('-----------')\n", " return None\n", " \n", "\n" ] }, { "cell_type": "code", "execution_count": 159, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "-----------\n", "regression coefficient : [[ 4.3904]]\n", "regression interception : [ 506.8]\n", "residues : [ 4004411.04653061]\n", "s2 = 80088.2209306\n", "\n", "confidence interval of s2: [58015.952325132173, 130205.6708168847]\n", "confidence interval of regression coefficient: [1.6017085525353174, 7.1790914474646854]\n", "confidence interval of regression interception: [344.97545415797902, 668.62454584202089]\n", "-----------\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "//anaconda/envs/g_dash/lib/python3.4/site-packages/sklearn/utils/deprecation.py:70: DeprecationWarning: Function residues_ is deprecated; ``residues_`` is deprecated and will be removed in 0.19\n", " warnings.warn(msg, category=DeprecationWarning)\n" ] }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEcCAYAAADKlrO6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "linear_model_(x,y,.95)" ] }, { "cell_type": "code", "execution_count": 162, "metadata": { "collapsed": true }, "outputs": [], "source": [ "# test again \n", "\n", "x2 = np.linspace(0,100,500).reshape(500,1)\n", "a = 15\n", "b = 200\n", "#noise = np.random.rand(100).reshape(100,1)\n", "noise = np.random.randint(1000, size=500).reshape(500,1)\n", "# y = 0.1*x + noise \n", "y2 = a*x2+noise" ] }, { "cell_type": "code", "execution_count": 163, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "-----------\n", "regression coefficient : [[ 13.99713868]]\n", "regression interception : [ 563.83906587]\n", "residues : [ 36160863.70569767]\n", "s2 = 72321.7274114\n", "\n", "confidence interval of s2: [64374.398826698103, 82547.719823322041]\n", "confidence interval of regression coefficient: [13.178586032136971, 14.815691333132476]\n", "confidence interval of regression interception: [516.55623555179125, 611.12189618473633]\n", "-----------\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "//anaconda/envs/g_dash/lib/python3.4/site-packages/sklearn/utils/deprecation.py:70: DeprecationWarning: Function residues_ is deprecated; ``residues_`` is deprecated and will be removed in 0.19\n", " warnings.warn(msg, category=DeprecationWarning)\n" ] }, { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYIAAAEcCAYAAADKlrO6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n", "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "linear_model_(x2,y2,.95)" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.4.5" } }, "nbformat": 4, "nbformat_minor": 2 }