Last active
December 14, 2015 15:18
-
-
Save econpy/5106304 to your computer and use it in GitHub Desktop.
Google Domestic Trends: Using automotive buyer index to predict automotive financing index.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
{ | |
"metadata": { | |
"name": "Automotive Google Trends" | |
}, | |
"nbformat": 3, | |
"nbformat_minor": 0, | |
"worksheets": [ | |
{ | |
"cells": [ | |
{ | |
"cell_type": "code", | |
"collapsed": false, | |
"input": "from pandas import read_csv,DatetimeIndex,ols\nfrom urllib import urlopen", | |
"language": "python", | |
"metadata": {}, | |
"outputs": [], | |
"prompt_number": 1 | |
}, | |
{ | |
"cell_type": "code", | |
"collapsed": false, | |
"input": "def get_index(gindex, startdate=20040101):\n \"\"\"\n API wrapper for Google Domestic Trends data.\n https://www.google.com/finance/domestic_trends\n\n Available Indices:\n\n 'ADVERT', 'AIRTVL', 'AUTOBY', 'AUTOFI', 'AUTO', 'BIZIND', 'BNKRPT',\n 'COMLND', 'COMPUT', 'CONSTR', 'CRCARD', 'DURBLE', 'EDUCAT', 'INVEST',\n 'FINPLN', 'FURNTR', 'INSUR', 'JOBS', 'LUXURY', 'MOBILE', 'MTGE',\n 'RLEST', 'RENTAL', 'SHOP', 'TRAVEL', 'UNEMPL'\n\n \"\"\"\n base_url = 'http://www.google.com/finance/historical?q=GOOGLEINDEX_US:'\n full_url = '%s%s&output=csv&startdate=%s' % (base_url, gindex, startdate)\n dframe = read_csv(urlopen(full_url), index_col=0)\n dframe.index = DatetimeIndex(dframe.index)\n dframe = dframe.sort_index(0)\n for col in dframe.columns:\n if len(dframe[col].unique()) == 1:\n dframe.pop(col)\n if len(dframe.columns) == 1 and dframe.columns[0] == 'Close':\n dframe.columns = [gindex]\n return dframe[gindex]", | |
"language": "python", | |
"metadata": {}, | |
"outputs": [], | |
"prompt_number": 2 | |
}, | |
{ | |
"cell_type": "code", | |
"collapsed": false, | |
"input": "autobuyers = get_index('AUTOBY') # https://www.google.com/finance?q=GOOGLEINDEX_US:AUTOBY\nautofinancing = get_index('AUTOFI') # https://www.google.com/finance?q=GOOGLEINDEX_US:AUTOFI", | |
"language": "python", | |
"metadata": {}, | |
"outputs": [], | |
"prompt_number": 3 | |
}, | |
{ | |
"cell_type": "code", | |
"collapsed": false, | |
"input": "# Run OLS testing if searches for queries related to people looking to purchase a car\n# are predictive of searches for automotive financing.\nmodel = ols(y=autofinancing, x={'Automotive Buyers': autobuyers})\nprint model.summary", | |
"language": "python", | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"output_type": "stream", | |
"stream": "stdout", | |
"text": "\n-------------------------Summary of Regression Analysis-------------------------\n\nFormula: Y ~ <Automotive Buyers> + <intercept>\n\nNumber of Observations: 3351\nNumber of Degrees of Freedom: 2\n\nR-squared: 0.6429\nAdj R-squared: 0.6428\n\nRmse: 0.0648\n\nF-stat (1, 3349): 6028.2607, p-value: 0.0000\n\nDegrees of Freedom: model 1, resid 3349\n\n-----------------------Summary of Estimated Coefficients------------------------\n Variable Coef Std Err t-stat p-value CI 2.5% CI 97.5%\n--------------------------------------------------------------------------------\nAutomotive Buyers 0.8114 0.0105 77.64 0.0000 0.7909 0.8319\n intercept 0.2166 0.0094 23.11 0.0000 0.1982 0.2350\n---------------------------------End of Summary---------------------------------\n\n" | |
} | |
], | |
"prompt_number": 4 | |
}, | |
{ | |
"cell_type": "code", | |
"collapsed": false, | |
"input": "# Plot actual Y vs. predicted Y\npred = model.predict()\npred.plot(color='b')\nautofinancing.plot(color='g')", | |
"language": "python", | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"output_type": "pyout", | |
"prompt_number": 5, | |
"text": "<matplotlib.axes.AxesSubplot at 0x5576f50>" | |
}, | |
{ | |
"output_type": "display_data", | |
"png": "iVBORw0KGgoAAAANSUhEUgAAAXUAAAD9CAYAAABDaefJAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n", | |
"text": "<matplotlib.figure.Figure at 0x5576c90>" | |
} | |
], | |
"prompt_number": 5 | |
}, | |
{ | |
"cell_type": "code", | |
"collapsed": false, | |
"input": "# Plot the residual\nerr = model.resid\nerr.plot(color='r')", | |
"language": "python", | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"output_type": "pyout", | |
"prompt_number": 6, | |
"text": "<matplotlib.axes.AxesSubplot at 0x5595990>" | |
}, | |
{ | |
"output_type": "display_data", | |
"png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAD9CAYAAABQvqc9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nH4uG72ttLEpQAAAAAElFTkSuQmCC\n", | |
"text": "<matplotlib.figure.Figure at 0x4970850>" | |
} | |
], | |
"prompt_number": 6 | |
}, | |
{ | |
"cell_type": "code", | |
"collapsed": false, | |
"input": "# Test simple model of 1-period autocorrelation\nerr_t1 = err.tshift(1, freq='D')\nautocorr_model = ols(y=err, x={'err_t1': err_t1})\nprint autocorr_model.summary", | |
"language": "python", | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"output_type": "stream", | |
"stream": "stdout", | |
"text": "\n-------------------------Summary of Regression Analysis-------------------------\n\nFormula: Y ~ <err_t1> + <intercept>\n\nNumber of Observations: 3350\nNumber of Degrees of Freedom: 2\n\nR-squared: 0.9389\nAdj R-squared: 0.9389\n\nRmse: 0.0160\n\nF-stat (1, 3348): 51439.6180, p-value: 0.0000\n\nDegrees of Freedom: model 1, resid 3348\n\n-----------------------Summary of Estimated Coefficients------------------------\n Variable Coef Std Err t-stat p-value CI 2.5% CI 97.5%\n--------------------------------------------------------------------------------\n err_t1 0.9689 0.0043 226.80 0.0000 0.9606 0.9773\n intercept 0.0000 0.0003 0.06 0.9555 -0.0005 0.0006\n---------------------------------End of Summary---------------------------------\n\n" | |
} | |
], | |
"prompt_number": 7 | |
}, | |
{ | |
"cell_type": "code", | |
"collapsed": false, | |
"input": "# Test model with up to 14-period autocorrelation\nerr_terms = {}\nfor lag in xrange(1, 15):\n err_terms['err_%s' % lag] = err.tshift(lag, freq='D')\nautocorr_model = ols(y=err, x=err_terms)\nprint autocorr_model.summary", | |
"language": "python", | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"output_type": "stream", | |
"stream": "stdout", | |
"text": "\n-------------------------Summary of Regression Analysis-------------------------\n\nFormula: Y ~ <err_1> + <err_10> + <err_11> + <err_12> + <err_13> + <err_14>\n + <err_2> + <err_3> + <err_4> + <err_5> + <err_6> + <err_7> + <err_8>\n + <err_9> + <intercept>\n\nNumber of Observations: 3337\nNumber of Degrees of Freedom: 15\n\nR-squared: 0.9544\nAdj R-squared: 0.9542\n\nRmse: 0.0139\n\nF-stat (14, 3322): 4965.7113, p-value: 0.0000\n\nDegrees of Freedom: model 14, resid 3322\n\n-----------------------Summary of Estimated Coefficients------------------------\n Variable Coef Std Err t-stat p-value CI 2.5% CI 97.5%\n--------------------------------------------------------------------------------\n err_1 1.2159 0.0173 70.32 0.0000 1.1820 1.2498\n err_10 -0.0440 0.0272 -1.61 0.1067 -0.0973 0.0094\n err_11 0.0378 0.0272 1.39 0.1656 -0.0156 0.0912\n err_12 -0.0087 0.0272 -0.32 0.7494 -0.0620 0.0447\n err_13 -0.0877 0.0271 -3.24 0.0012 -0.1409 -0.0346\n--------------------------------------------------------------------------------\n err_14 0.0724 0.0172 4.20 0.0000 0.0386 0.1061\n err_2 -0.1353 0.0272 -4.97 0.0000 -0.1887 -0.0820\n err_3 -0.0437 0.0273 -1.60 0.1096 -0.0972 0.0098\n err_4 -0.0123 0.0273 -0.45 0.6536 -0.0658 0.0413\n err_5 -0.0436 0.0272 -1.60 0.1094 -0.0970 0.0098\n--------------------------------------------------------------------------------\n err_6 -0.0504 0.0272 -1.85 0.0642 -0.1038 0.0030\n err_7 -0.3302 0.0262 -12.61 0.0000 -0.3815 -0.2789\n err_8 0.4320 0.0262 16.49 0.0000 0.3806 0.4833\n err_9 -0.0325 0.0272 -1.19 0.2329 -0.0859 0.0209\n intercept 0.0000 0.0002 0.03 0.9755 -0.0005 0.0005\n---------------------------------End of Summary---------------------------------\n\n" | |
} | |
], | |
"prompt_number": 8 | |
}, | |
{ | |
"cell_type": "code", | |
"collapsed": false, | |
"input": "# Find the lagged error terms that are significant at the 95% level\nsignificantvals = []\nfor pval, term, paramval in zip(autocorr_model.p_value, autocorr_model.beta.index, autocorr_model.beta):\n if pval < 0.05:\n significantvals.append((term, paramval))\n\n# Sort them by the size of the coefficient\nfrom operator import itemgetter\nsorted_significantvals = sorted(significantvals, key=itemgetter(1))\nfor sv in sorted_significantvals:\n print '%s: %s' % (sv[0], sv[1])", | |
"language": "python", | |
"metadata": {}, | |
"outputs": [ | |
{ | |
"output_type": "stream", | |
"stream": "stdout", | |
"text": "err_7: -0.330199121438\nerr_2: -0.135321015861\nerr_13: -0.087741039288\nerr_14: 0.0723747974828\nerr_8: 0.43198713133\nerr_1: 1.21586269622\n" | |
} | |
], | |
"prompt_number": 9 | |
}, | |
{ | |
"cell_type": "raw", | |
"metadata": {}, | |
"source": " " | |
} | |
], | |
"metadata": {} | |
} | |
] | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment