└── A_B_test.ipynb /A_B_test.ipynb: -------------------------------------------------------------------------------- 1 | { 2 | "cells": [ 3 | { 4 | "cell_type": "code", 5 | "execution_count": 172, 6 | "metadata": {}, 7 | "outputs": [], 8 | "source": [ 9 | "import numpy as np\n", 10 | "import pandas as pd\n", 11 | "import random\n", 12 | "import matplotlib.pyplot as plt\n", 13 | "import seaborn as sns\n", 14 | "import statsmodels.api as sm\n", 15 | "\n", 16 | "np.random.seed(100)" 17 | ] 18 | }, 19 | { 20 | "cell_type": "code", 21 | "execution_count": 173, 22 | "metadata": {}, 23 | "outputs": [], 24 | "source": [ 25 | "# Run 10,000 experiments to examine the distribution of the differences between control and experimental groups\n", 26 | "# Both control and experimental groups are random normal variables\n", 27 | "\n", 28 | "sims = 10000\n", 29 | "mc_1 = []\n", 30 | "mc_2 = []\n", 31 | "mean_diff = []\n", 32 | "\n", 33 | "for sim in range(sims):\n", 34 | " # Control\n", 35 | " N = 500\n", 36 | " mean_1 = 0.12\n", 37 | " stdev_1 = 0.05\n", 38 | " group_1 = []\n", 39 | " [group_1.append(np.random.randn()*stdev_1 + mean_1) for i in range(N)];\n", 40 | "\n", 41 | " # Experimental\n", 42 | " N = 500\n", 43 | " mean_2 = 0.12\n", 44 | " stdev_2 = 0.05\n", 45 | " group_2 = []\n", 46 | " [group_2.append(np.random.randn()*stdev_2 + mean_2) for i in range(N)];\n", 47 | " \n", 48 | " mc_1.append(np.mean(group_1))\n", 49 | " mc_2.append(np.mean(group_2))\n", 50 | " mean_diff.append(np.mean(group_2) - np.mean(group_1))" 51 | ] 52 | }, 53 | { 54 | "cell_type": "code", 55 | "execution_count": 174, 56 | "metadata": {}, 57 | "outputs": [ 58 | { 59 | "data": { 60 | "image/png": "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\n", 61 | "text/plain": [ 62 | "
" 63 | ] 64 | }, 65 | "metadata": { 66 | "needs_background": "light" 67 | }, 68 | "output_type": "display_data" 69 | } 70 | ], 71 | "source": [ 72 | "# Plot histogram of savings rates for 1 sample\n", 73 | "\n", 74 | "fig, ax = plt.subplots(figsize=(12,8))\n", 75 | "sns.distplot(group_1, kde=False, bins=25, label='Control');\n", 76 | "sns.distplot(group_2, kde=False, bins=25, label='Experimental');\n", 77 | "\n", 78 | "ax.set_xlabel(\"Savings Rate\",fontsize=16)\n", 79 | "ax.set_ylabel(\"Frequency\",fontsize=16)\n", 80 | "plt.legend()\n", 81 | "plt.setp(ax.get_legend().get_texts(), fontsize=16)\n", 82 | "plt.tight_layout()\n", 83 | "\n", 84 | "plt.savefig(fname='hist_1', dpi=150)\n", 85 | "plt.show()" 86 | ] 87 | }, 88 | { 89 | "cell_type": "code", 90 | "execution_count": 175, 91 | "metadata": {}, 92 | "outputs": [ 93 | { 94 | "data": { 95 | "image/png": "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\n", 96 | "text/plain": [ 97 | "
" 98 | ] 99 | }, 100 | "metadata": { 101 | "needs_background": "light" 102 | }, 103 | "output_type": "display_data" 104 | } 105 | ], 106 | "source": [ 107 | "# Plot histogram of results for all 10,000 experiments\n", 108 | "\n", 109 | "fig, ax = plt.subplots(figsize=(12,8))\n", 110 | "sns.distplot(mean_diff, kde=False, bins=50, label='Difference of Means');\n", 111 | "\n", 112 | "ax.set_xlabel(\"Difference of Means\",fontsize=16)\n", 113 | "ax.set_ylabel(\"Frequency\",fontsize=16)\n", 114 | "plt.axvline(x=0.01, c='red')\n", 115 | "plt.legend()\n", 116 | "plt.setp(ax.get_legend().get_texts(), fontsize=16)\n", 117 | "plt.tight_layout()\n", 118 | "\n", 119 | "plt.savefig(fname='hist_2', dpi=150)\n", 120 | "plt.show()" 121 | ] 122 | }, 123 | { 124 | "cell_type": "code", 125 | "execution_count": 176, 126 | "metadata": {}, 127 | "outputs": [ 128 | { 129 | "data": { 130 | "image/png": "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\n", 131 | "text/plain": [ 132 | "
" 133 | ] 134 | }, 135 | "metadata": { 136 | "needs_background": "light" 137 | }, 138 | "output_type": "display_data" 139 | } 140 | ], 141 | "source": [ 142 | "# Q-Q plot to check for normal distribution\n", 143 | "\n", 144 | "fig_qq, ax_qq = plt.subplots(figsize=(5,5))\n", 145 | "sm.qqplot(np.array(mean_diff), fit=True, line='45', ax=ax_qq);\n", 146 | "plt.savefig(fname='qq_plot', dpi=150)" 147 | ] 148 | }, 149 | { 150 | "cell_type": "code", 151 | "execution_count": 177, 152 | "metadata": {}, 153 | "outputs": [ 154 | { 155 | "name": "stdout", 156 | "output_type": "stream", 157 | "text": [ 158 | "Number of obs. >= experimental result: 9\n", 159 | "Percentage of obs. >= experimental result: 0.0009\n" 160 | ] 161 | } 162 | ], 163 | "source": [ 164 | "print('Number of obs. >= experimental result: ' + str(sum([1 for i in mean_diff if i >= 0.01])))\n", 165 | "print('Percentage of obs. >= experimental result: ' + str(sum([1 for i in mean_diff if i >= 0.01])/sims))" 166 | ] 167 | }, 168 | { 169 | "cell_type": "code", 170 | "execution_count": 178, 171 | "metadata": {}, 172 | "outputs": [ 173 | { 174 | "name": "stdout", 175 | "output_type": "stream", 176 | "text": [ 177 | "Stdev of mean diff from experiment: 0.00315\n" 178 | ] 179 | } 180 | ], 181 | "source": [ 182 | "print('Stdev of mean diff from experiment: ' + str(round(np.var(mean_diff)**0.5, 5)))" 183 | ] 184 | }, 185 | { 186 | "cell_type": "code", 187 | "execution_count": 179, 188 | "metadata": {}, 189 | "outputs": [ 190 | { 191 | "name": "stdout", 192 | "output_type": "stream", 193 | "text": [ 194 | "Stdev of mean diff analytical: 0.00312\n" 195 | ] 196 | } 197 | ], 198 | "source": [ 199 | "print('Stdev of mean diff analytical: ' + str(round((np.var(group_1)/len(group_1) + np.var(group_2)/len(group_2))**0.5, 5)))" 200 | ] 201 | }, 202 | { 203 | "cell_type": "code", 204 | "execution_count": 180, 205 | "metadata": {}, 206 | "outputs": [ 207 | { 208 | "name": "stdout", 209 | "output_type": "stream", 210 | "text": [ 211 | "Standard Error: 0.0031622776601683794\n" 212 | ] 213 | } 214 | ], 215 | "source": [ 216 | "print('Standard Error: ' + str((2*0.0025/500)**0.5))" 217 | ] 218 | }, 219 | { 220 | "cell_type": "code", 221 | "execution_count": 181, 222 | "metadata": {}, 223 | "outputs": [ 224 | { 225 | "name": "stdout", 226 | "output_type": "stream", 227 | "text": [ 228 | "Test Statistic: 3.1622776601683795\n" 229 | ] 230 | } 231 | ], 232 | "source": [ 233 | "print('Test Statistic: ' + str((0.01 - 0)/(2*0.0025/500)**0.5))" 234 | ] 235 | }, 236 | { 237 | "cell_type": "code", 238 | "execution_count": 182, 239 | "metadata": {}, 240 | "outputs": [ 241 | { 242 | "name": "stdout", 243 | "output_type": "stream", 244 | "text": [ 245 | "The p-value is: 0.0016\n" 246 | ] 247 | } 248 | ], 249 | "source": [ 250 | "from scipy.stats import norm\n", 251 | "\n", 252 | "# Two Tailed Test\n", 253 | "print('The p-value is: ' + str(round((1 - norm.cdf(3.16))*2,4)))" 254 | ] 255 | }, 256 | { 257 | "cell_type": "code", 258 | "execution_count": null, 259 | "metadata": {}, 260 | "outputs": [], 261 | "source": [] 262 | } 263 | ], 264 | "metadata": { 265 | "kernelspec": { 266 | "display_name": "Python 3", 267 | "language": "python", 268 | "name": "python3" 269 | }, 270 | "language_info": { 271 | "codemirror_mode": { 272 | "name": "ipython", 273 | "version": 3 274 | }, 275 | "file_extension": ".py", 276 | "mimetype": "text/x-python", 277 | "name": "python", 278 | "nbconvert_exporter": "python", 279 | "pygments_lexer": "ipython3", 280 | "version": "3.7.1" 281 | } 282 | }, 283 | "nbformat": 4, 284 | "nbformat_minor": 2 285 | } 286 | --------------------------------------------------------------------------------