{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Correlation
"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"A correlation is the degree of association between two variables. One of my favorite books on this topic is: \n",
"\n",
"\n",
"\n",
"and they illustrate it by looking at\n",
"## Ladies expenditures on clothes and makeup\n",
"\n",
"\n",
"\n",
"\n",
"So let's go ahead and create that data in Pandas and show the table:"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" clothes | \n",
" makeup | \n",
"
\n",
" \n",
" \n",
" \n",
" Ms A | \n",
" 7000 | \n",
" 3000 | \n",
"
\n",
" \n",
" Ms B | \n",
" 8000 | \n",
" 5000 | \n",
"
\n",
" \n",
" Ms C | \n",
" 25000 | \n",
" 12000 | \n",
"
\n",
" \n",
" Ms D | \n",
" 5000 | \n",
" 2000 | \n",
"
\n",
" \n",
" Ms E | \n",
" 12000 | \n",
" 7000 | \n",
"
\n",
" \n",
" Ms F | \n",
" 30000 | \n",
" 15000 | \n",
"
\n",
" \n",
" Ms G | \n",
" 10000 | \n",
" 5000 | \n",
"
\n",
" \n",
" Ms H | \n",
" 15000 | \n",
" 6000 | \n",
"
\n",
" \n",
" Ms I | \n",
" 20000 | \n",
" 8000 | \n",
"
\n",
" \n",
" Ms J | \n",
" 18000 | \n",
" 10000 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" clothes makeup\n",
"Ms A 7000 3000\n",
"Ms B 8000 5000\n",
"Ms C 25000 12000\n",
"Ms D 5000 2000\n",
"Ms E 12000 7000\n",
"Ms F 30000 15000\n",
"Ms G 10000 5000\n",
"Ms H 15000 6000\n",
"Ms I 20000 8000\n",
"Ms J 18000 10000"
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"import pandas as pd\n",
"import numpy as np\n",
"from pandas import DataFrame\n",
"makeup = [3000, 5000, 12000, 2000, 7000, 15000, 5000, 6000, 8000, 10000]\n",
"clothes = [7000, 8000, 25000, 5000, 12000, 30000, 10000, 15000, 20000, 18000]\n",
"ladies = ['Ms A','Ms B','Ms C','Ms D','Ms E','Ms F','Ms G','Ms H','Ms I','Ms J',]\n",
"monthly = DataFrame({'makeup': makeup, 'clothes': clothes}, index= ladies)\n",
"monthly\n",
" \n",
"\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"and let's show the scatterplot"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAYcAAAD8CAYAAACcjGjIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAFtZJREFUeJzt3W2MnfWZ3/HvtbZLZjeB4WHWtcdO7SyuJShaHI6Qt46q\nNHR33Gi1dhBKveoWr5bCSqAoaSNXnkTq7r7C1LuhQlXYsiXFUDbgEmdAYVmX2JG2L4rpOEMyPE2Z\nFig+GOwFBm/bEbKdqy/O/8CZuR3m6Yxnzvj7kY7Of677/p9zHbD9m/vp3JGZSJLU6hcWugFJ0uJj\nOEiSKgwHSVKF4SBJqjAcJEkVhoMkqcJwkCRVGA6SpArDQZJUsXyhG5itK664ItetW7fQbUhSRzl6\n9OhfZ2bPVOt1bDisW7eOwcHBhW5DkjpKRLw+nfXcrSRJqjAcJEkVhoMkqcJwkCRVGA6SpIopwyEi\nPhERz0bETyLihYj4o1K/LCKejohXyvOlLXP6I2I0IkYioq+lfl1EDJdl90RElPpFEfFoqR+JiHXt\n/6iS1NkGhups2XOY9bufZMuewwwM1eftvaaz5fAB8IXM/FXgWmBrRGwGdgOHMnMDcKj8TERcBewA\nrga2At+OiGXlte4FbgU2lMfWUr8FeC8zrwTuBu5qw2eTpCVjYKhO/4Fh6mPjJFAfG6f/wPC8BcSU\n4ZAN/6f8uKI8EtgG7Cv1fcD2Mt4GPJKZH2Tmq8AocH1ErAIuzsxnsnFv0gcnzWm+1mPADc2tCkkS\n7D04wvjpsxNq46fPsvfgyLy837SOOUTEsoh4DjgBPJ2ZR4CVmXm8rPIWsLKMe4E3WqYfK7XeMp5c\nnzAnM88A7wOXn6OP2yJiMCIGT548OZ3WJWlJeHNsfEb1uZpWOGTm2cy8FlhDYyvg701anjS2JuZV\nZt6XmbXMrPX0THn1tyQtGau7u2ZUn6sZna2UmWPAj2gcK3i77CqiPJ8oq9WBtS3T1pRavYwn1yfM\niYjlwCXAOzPpTZKWsl19G+lasWxCrWvFMnb1bZyX95vO2Uo9EdFdxl3ArwMvA08AO8tqO4HHy/gJ\nYEc5A2k9jQPPz5ZdUKciYnM5nnDzpDnN17oJOFy2RiRJwPZNvdx54zX0dncRQG93F3feeA3bN/VO\nOXc2pvPFe6uAfeWMo18A9mfmDyLivwH7I+IW4HXgywCZ+UJE7AdeBM4Ad2Rm8yjK7cADQBfwVHkA\n3A88FBGjwLs0znaSJLXYvql33sJgsujUX9BrtVr6raySNDMRcTQza1Ot5xXSkqQKw0GSVGE4SJIq\nDAdJUoXhIEmqMBwkSRWGgySpwnCQJFUYDpKkCsNBklRhOEiSKgwHSVKF4SBJqjAcJEkVhoMkqcJw\nkCRVGA6SpArDQZJUYThIkioMB0lSheEgSaowHCRJFYaDJKnCcJAkVRgOkqQKw0GSVGE4SJIqDAdJ\nUoXhIEmqMBwkSRVThkNErI2IH0XEixHxQkR8tdT/MCLqEfFceXyxZU5/RIxGxEhE9LXUr4uI4bLs\nnoiIUr8oIh4t9SMRsa79H1WSNF3Lp7HOGeDrmfnjiPgUcDQini7L7s7MP25dOSKuAnYAVwOrgR9G\nxN/NzLPAvcCtwBHgL4CtwFPALcB7mXllROwA7gL+ydw/nqROMzBUZ+/BEd4cG2d1dxe7+jayfVPv\nQrd1wZlyyyEzj2fmj8v4b4CXgI/7P7UNeCQzP8jMV4FR4PqIWAVcnJnPZGYCDwLbW+bsK+PHgBua\nWxWSLhwDQ3X6DwxTHxsngfrYOP0HhhkYqi90axecGR1zKLt7NtH4zR/gKxHx04j4TkRcWmq9wBst\n046VWm8ZT65PmJOZZ4D3gctn0pukzrf34Ajjp89OqI2fPsvegyML1NGFa9rhEBGfBL4HfC0zT9HY\nRfQZ4FrgOPAn89LhxB5ui4jBiBg8efLkfL+dpPPszbHxGdU1f6YVDhGxgkYwPJyZBwAy8+3MPJuZ\nPwP+DLi+rF4H1rZMX1Nq9TKeXJ8wJyKWA5cA70zuIzPvy8xaZtZ6enqm9wkldYzV3V0zqmv+TOds\npQDuB17KzG+11Fe1rPYl4PkyfgLYUc5AWg9sAJ7NzOPAqYjYXF7zZuDxljk7y/gm4HA5LiHpArKr\nbyNdK5ZNqHWtWMauvo0L1NGFazpnK20B/hkwHBHPldo3gN+OiGuBBF4Dfh8gM1+IiP3AizTOdLqj\nnKkEcDvwANBF4yylp0r9fuChiBgF3qVxtpOkC0zzrCTPVlp40am/oNdqtRwcHFzoNiSpo0TE0cys\nTbWeV0hLkioMB0lSheEgSaowHCRJFYaDJKnCcJAkVRgOkqQKw0GSVGE4SJIqDAdJUoXhIEmqMBwk\nSRWGgySpwnCQJFUYDpKkCsNBklRhOEiSKgwHSVKF4SBJqjAcJEkVhoMkqcJwkCRVGA6SpArDQZJU\nYThIkioMB0lSheEgSaowHCRJFYaDJKnCcJAkVRgOkqSKKcMhItZGxI8i4sWIeCEivlrql0XE0xHx\nSnm+tGVOf0SMRsRIRPS11K+LiOGy7J6IiFK/KCIeLfUjEbGu/R9VmrmBoTpb9hxm/e4n2bLnMAND\n9YVuSTovprPlcAb4emZeBWwG7oiIq4DdwKHM3AAcKj9Tlu0Arga2At+OiGXlte4FbgU2lMfWUr8F\neC8zrwTuBu5qw2eT5mRgqE7/gWHqY+MkUB8bp//AsAGhC8KU4ZCZxzPzx2X8N8BLQC+wDdhXVtsH\nbC/jbcAjmflBZr4KjALXR8Qq4OLMfCYzE3hw0pzmaz0G3NDcqpAWyt6DI4yfPjuhNn76LHsPjixQ\nR9L5M6NjDmV3zybgCLAyM4+XRW8BK8u4F3ijZdqxUust48n1CXMy8wzwPnD5Od7/togYjIjBkydP\nzqR1acbeHBufUV1aSqYdDhHxSeB7wNcy81TrsrIlkG3urSIz78vMWmbWenp65vvtdIFb3d01o7q0\nlEwrHCJiBY1geDgzD5Ty22VXEeX5RKnXgbUt09eUWr2MJ9cnzImI5cAlwDsz/TBSO+3q20jXimUT\nal0rlrGrb+MCdSSdP9M5WymA+4GXMvNbLYueAHaW8U7g8Zb6jnIG0noaB56fLbugTkXE5vKaN0+a\n03ytm4DDZWtEWjDbN/Vy543X0NvdRQC93V3ceeM1bN/UO+VcqdPFVP8GR8TngP8KDAM/K+Vv0Dju\nsB/4NPA68OXMfLfM+SbwezTOdPpaZj5V6jXgAaALeAr4SmZmRHwCeIjG8Yx3gR2Z+b8+rq9arZaD\ng4Mz/bySdEGLiKOZWZtyvU79Bd1wkKSZm244eIW0JKnCcJAkVRgOkqQKw0GSVGE4SJIqDAdJUoXh\nIEmqWL7QDUiaHwNDdfYeHOHNsXFWd3exq2+jV3dr2gwHaQlq3oui+ZXjzXtRAAaEpsXdStIS5L0o\nNFeGg7QEeS8KzZXhIC1B3otCc2U4SEuQ96LQXHlAWlqCmgedPVtJs2U4SEvU9k29hoFmzd1KkqQK\nw0GSVGE4SJIqDAdJUoXhIEmqMBwkSRWGgySpwnCQJFUYDpKkCsNBklRhOEiSKgwHSVKF4SBJqjAc\nJEkVhoMkqWLKcIiI70TEiYh4vqX2hxFRj4jnyuOLLcv6I2I0IkYioq+lfl1EDJdl90RElPpFEfFo\nqR+JiHXt/YiSpJmazpbDA8DWc9Tvzsxry+MvACLiKmAHcHWZ8+2IaN6r8F7gVmBDeTRf8xbgvcy8\nErgbuGuWn0WL1MBQnS17DrN+95Ns2XOYgaH6QrckaQpThkNm/hXw7jRfbxvwSGZ+kJmvAqPA9RGx\nCrg4M5/JzAQeBLa3zNlXxo8BNzS3KtT5Bobq9B8Ypj42TgL1sXH6DwwbENIiN5djDl+JiJ+W3U6X\nllov8EbLOsdKrbeMJ9cnzMnMM8D7wOVz6EuLyN6DI4yfPjuhNn76LHsPjixQR5KmY7bhcC/wGeBa\n4DjwJ23r6GNExG0RMRgRgydPnjwfb6k5enNsfEZ1SYvDrMIhM9/OzLOZ+TPgz4Dry6I6sLZl1TWl\nVi/jyfUJcyJiOXAJ8M7Ped/7MrOWmbWenp7ZtK7zbHV314zqkhaHWYVDOYbQ9CWgeSbTE8COcgbS\nehoHnp/NzOPAqYjYXI4n3Aw83jJnZxnfBBwuxyW0BOzq20jXimUTal0rlrGrb+MCdSRpOpZPtUJE\nfBf4PHBFRBwD/gD4fERcCyTwGvD7AJn5QkTsB14EzgB3ZGZzh/PtNM586gKeKg+A+4GHImKUxoHv\nHe34YFoctm9qHFrae3CEN8fGWd3dxa6+jR/WJS1O0am/pNdqtRwcHFzoNiSpo0TE0cysTbWeV0hL\nkioMB0lSheEgSaowHCRJFYaDJKnCcJAkVRgOkqQKw0GSVGE4SJIqDAdJUoXhIEmqMBwkSRWGgySp\nwnCQJFUYDpKkCsNBklQx5Z3gpMVsYKjuXeakeWA4qGMNDNXpPzDM+OnGnWjrY+P0HxgGMCCkOXK3\nkjrW3oMjHwZD0/jps+w9OLJAHUlLh+GgjvXm2PiM6pKmz3BQx1rd3TWjuqTpMxzUsXb1baRrxbIJ\nta4Vy9jVt3GBOpKWDg9Iq2M1Dzp7tpLUfoaDOtr2Tb2GgTQP3K0kSaowHCRJFYaDJKnCcJAkVRgO\nkqQKw0GSVGE4SJIqpgyHiPhORJyIiOdbapdFxNMR8Up5vrRlWX9EjEbESET0tdSvi4jhsuyeiIhS\nvygiHi31IxGxrr0fUdMxMFRny57DrN/9JFv2HGZgqL7QLUlaQNPZcngA2Dqpths4lJkbgEPlZyLi\nKmAHcHWZ8+2IaH6/wb3ArcCG8mi+5i3Ae5l5JXA3cNdsP4xmp/nV1/WxcZKPvvragJAuXFOGQ2b+\nFfDupPI2YF8Z7wO2t9QfycwPMvNVYBS4PiJWARdn5jOZmcCDk+Y0X+sx4IbmVoXOD7/6WtJksz3m\nsDIzj5fxW8DKMu4F3mhZ71ip9Zbx5PqEOZl5BngfuPxcbxoRt0XEYEQMnjx5cpatazK/+lrSZHM+\nIF22BLINvUznve7LzFpm1np6es7HW14Q/OprSZPNNhzeLruKKM8nSr0OrG1Zb02p1ct4cn3CnIhY\nDlwCvDPLvjQLfvW1pMlmGw5PADvLeCfweEt9RzkDaT2NA8/Pll1QpyJiczmecPOkOc3Xugk4XLZG\ndJ5s39TLnTdeQ293FwH0dndx543X+G2n0gVsyq/sjojvAp8HroiIY8AfAHuA/RFxC/A68GWAzHwh\nIvYDLwJngDsys3mk83YaZz51AU+VB8D9wEMRMUrjwPeOtnwyzYhffS2pVXTqL+m1Wi0HBwcXug1J\n6igRcTQza1Ot5xXSkqQKw0GSVGE4SJIqDAdJUoXhIEmqMBwkSRWGgySpYsqL4HRhGBiqs/fgCG+O\njbO6u4tdfRu9KE66gBkO+vB+Ds2v7W7ezwEwIKQLlLuV5P0cJFUYDvJ+DpIqDAd5PwdJFYaDvJ+D\npAoPSOvDg86erSSpyXAQ4P0cJE1kOLSR1wpIWioMhzbxWgFJS4kHpNvEawUkLSWGQ5t4rYCkpcRw\naBOvFZC0lBgObeK1ApKWEg9It4nXCkhaSgyHNvJaAUlLhbuVJEkVhoMkqcJwkCRVGA6SpArDQZJU\nYThIkioMB0lSxZzCISJei4jhiHguIgZL7bKIeDoiXinPl7as3x8RoxExEhF9LfXryuuMRsQ9ERFz\n6UuSNDft2HL4h5l5bWbWys+7gUOZuQE4VH4mIq4CdgBXA1uBb0dE8/sm7gVuBTaUx9Y29FUxMFRn\ny57DrN/9JFv2HGZgqD4fbyNJHW8+dittA/aV8T5ge0v9kcz8IDNfBUaB6yNiFXBxZj6TmQk82DKn\nbZr3W6iPjZN8dL8FA0KSquYaDgn8MCKORsRtpbYyM4+X8VvAyjLuBd5omXus1HrLeHK9rbzfgiRN\n31y/W+lzmVmPiF8Gno6Il1sXZmZGRM7xPT5UAug2gE9/+tMzmuv9FiRp+ua05ZCZ9fJ8Avg+cD3w\ndtlVRHk+UVavA2tbpq8ptXoZT66f6/3uy8xaZtZ6enpm1Kv3W5Ck6Zt1OETEL0XEp5pj4DeA54En\ngJ1ltZ3A42X8BLAjIi6KiPU0Djw/W3ZBnYqIzeUspZtb5rSN91uQpOmby26llcD3y1mny4E/z8y/\njIj/DuyPiFuA14EvA2TmCxGxH3gROAPckZnNgwC3Aw8AXcBT5dFW3m9BkqYvGicIdZ5arZaDg4ML\n3YYkdZSIONpy6cHP5RXSkqQKw0GSVGE4SJIqDAdJUoXhIEmq6NizlSLiJI1TZWfjCuCv29jO+WTv\nC8Pez79O7RsWd+9/JzOnvIq4Y8NhLiJicDqnci1G9r4w7P3869S+obN7b3K3kiSpwnCQJFVcqOFw\n30I3MAf2vjDs/fzr1L6hs3sHLtBjDpKkj3ehbjlIkj7GkgiHiFgbET+KiBcj4oWI+GqpXxYRT0fE\nK+X50pY5/RExGhEjEdHXUr8uIobLsnvK14ifj8+wLCKGIuIHndR7RHRHxGMR8XJEvBQRv9ZBvf+L\n8ufl+Yj4bkR8YrH2HhHfiYgTEfF8S61tvZav0n+01I9ExLp57n1v+TPz04j4fkR0L7bez9V3y7Kv\nR0RGxBWLre+2ycyOfwCrgM+W8aeA/wFcBfwbYHep7wbuKuOrgJ8AFwHrgf8JLCvLngU2A0Hjq8P/\n8Xn6DP8S+HPgB+Xnjuidxn3C/3kZ/y2guxN6p3Er2leBrvLzfuB3F2vvwD8APgs831JrW680vjb/\nT8t4B/DoPPf+G8DyMr5rMfZ+rr5LfS1wkMZ1Vlcstr7b9v9toRuYlw/VuFnQrwMjwKpSWwWMlHE/\n0N+y/kHg18o6L7fUfxv49+eh3zXAIeALfBQOi7534BIa/8DGpHon9N68p/llNO5H8oPyD9ai7R1Y\nx8R/YNvWa3OdMl5O4wKumK/eJy37EvDwYuz9XH0DjwG/CrzGR+GwqPpux2NJ7FZqVTbNNgFHgJXZ\nuNMcwFs0blAEH/3D0HSs1HrLeHJ9vv1b4F8BP2updULv64GTwH8su8T+QzTuCrjoe8/GLW7/GPjf\nwHHg/cz8L3RA7y3a2euHczLzDPA+cPn8tF3xe3x0g69F3XtEbAPqmfmTSYsWdd+zsaTCISI+CXwP\n+Fpmnmpdlo14XnSnZkXEbwInMvPoz1tnsfZO47edzwL3ZuYm4P/S2L3xocXae9k/v41GwK0Gfiki\nfqd1ncXa+7l0Uq+tIuKbNO4M+fBC9zKViPhF4BvAv17oXs6HJRMOEbGCRjA8nJkHSvntiFhVlq8C\nTpR6ncZ+w6Y1pVYv48n1+bQF+K2IeA14BPhCRPwnOqP3Y8CxzDxSfn6MRlh0Qu//CHg1M09m5mng\nAPD3O6T3pnb2+uGciFhOY5fhO/PWeeN9fhf4TeCflnCb0MekHhdD779C45eJn5S/r2uAH0fE317k\nfc/KkgiHcvT/fuClzPxWy6IngJ1lvJPGsYhmfUc5W2A9sAF4tmyin4qIzeU1b26ZMy8ysz8z12Tm\nOhoHpQ5n5u90SO9vAW9ExMZSuoHGPcIXfe80didtjohfLO95A/BSh/Te1M5eW1/rJhp/DudtSyQi\nttLYlfpbmfn/Jn2mRdl7Zg5n5i9n5rry9/UYjRNh3lrMfc/aQh/0aMcD+ByNTeqfAs+Vxxdp7L87\nBLwC/BC4rGXON2mcUTBCy9klQA14viz7d5zHA0TA5/nogHRH9A5cCwyW//YDwKUd1PsfAS+X932I\nxpkmi7J34Ls0jo2cpvGP0i3t7BX4BPCfgVEaZ9d8Zp57H6Wxv7359/VPF1vv5+p70vLXKAekF1Pf\n7Xp4hbQkqWJJ7FaSJLWX4SBJqjAcJEkVhoMkqcJwkCRVGA6SpArDQZJUYThIkir+PzRofEzkatT/\nAAAAAElFTkSuQmCC\n",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Import package\n",
"import matplotlib.pyplot as plt\n",
"\n",
"# Build Scatter plot\n",
"plt.scatter(monthly['makeup'], monthly['clothes'])\n",
"\n",
"\n",
"# Show plot\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"That's pretty good. But notice in the picture of the chart from The Manga Guide to Statistics that there is a title to the chart as well as labels for the x and y axis. Can you add those to the chart?\n",
"\n",
"\n",
"Task 1: labels
\n",
"\n"
]
},
{
"cell_type": "code",
"execution_count": 27,
"metadata": {},
"outputs": [],
"source": [
"# TBD"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"When the data points are close to a straight line going up, we say that there is a positive correlation between the two variables. It's sort of like saying:\n",
"\n",
" Ann: Is there any relationship between how much a person spends on makeup and how much they spend on clothes?\n",
" Ms. Graph: Yes. The more a person spends on makeup the more she spends on clothes.\n",
"\n",
"Let's look at a few more examples:\n",
"\n",
"## Weight and calories in 1-3 yr/old children\n",
"This is real data on the weight and calories consumed for children between 1 and 3 years of age."
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" calories | \n",
" weight | \n",
"
\n",
" \n",
" \n",
" \n",
" 0 | \n",
" 360 | \n",
" 7.7 | \n",
"
\n",
" \n",
" 1 | \n",
" 400 | \n",
" 7.8 | \n",
"
\n",
" \n",
" 2 | \n",
" 500 | \n",
" 8.6 | \n",
"
\n",
" \n",
" 3 | \n",
" 370 | \n",
" 8.5 | \n",
"
\n",
" \n",
" 4 | \n",
" 525 | \n",
" 8.6 | \n",
"
\n",
" \n",
" 5 | \n",
" 800 | \n",
" 9.0 | \n",
"
\n",
" \n",
" 6 | \n",
" 900 | \n",
" 10.1 | \n",
"
\n",
" \n",
" 7 | \n",
" 1200 | \n",
" 11.5 | \n",
"
\n",
" \n",
" 8 | \n",
" 1000 | \n",
" 11.0 | \n",
"
\n",
" \n",
" 9 | \n",
" 1400 | \n",
" 10.2 | \n",
"
\n",
" \n",
" 10 | \n",
" 1600 | \n",
" 11.9 | \n",
"
\n",
" \n",
" 11 | \n",
" 850 | \n",
" 10.4 | \n",
"
\n",
" \n",
" 12 | \n",
" 575 | \n",
" 9.3 | \n",
"
\n",
" \n",
" 13 | \n",
" 425 | \n",
" 9.1 | \n",
"
\n",
" \n",
" 14 | \n",
" 950 | \n",
" 8.5 | \n",
"
\n",
" \n",
" 15 | \n",
" 800 | \n",
" 11.0 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" calories weight\n",
"0 360 7.7\n",
"1 400 7.8\n",
"2 500 8.6\n",
"3 370 8.5\n",
"4 525 8.6\n",
"5 800 9.0\n",
"6 900 10.1\n",
"7 1200 11.5\n",
"8 1000 11.0\n",
"9 1400 10.2\n",
"10 1600 11.9\n",
"11 850 10.4\n",
"12 575 9.3\n",
"13 425 9.1\n",
"14 950 8.5\n",
"15 800 11.0"
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"weight = [7.7, 7.8, 8.6, 8.5, 8.6, 9, 10.1, 11.5, 11, 10.2, 11.9, 10.4, 9.3, 9.1, 8.5, 11]\n",
"calories = [360, 400, 500, 370, 525, 800, 900, 1200, 1000, 1400, 1600, 850, 575, 425, 950, 800]\n",
"kids = DataFrame({'weight': weight, 'calories': calories})\n",
"kids"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"\n",
"Task 2: draw the scatter plot with appropriate labels
\n",
"We are interested in seeing the correlation between weight and calories consumed.\n",
"\n"
]
},
{
"cell_type": "code",
"execution_count": 28,
"metadata": {},
"outputs": [],
"source": [
"# Your code here"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": true
},
"source": [
"What do you think? Is there a correlation?
\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"## The stronger the correlation the closer to a straight line:\n",
"The more the datapoints are in a straight line, the sronger the correlation. Let's look at students' midtern and final grades.\n"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" final | \n",
" midterm | \n",
"
\n",
" \n",
" \n",
" \n",
" 0 | \n",
" 72 | \n",
" 65 | \n",
"
\n",
" \n",
" 1 | \n",
" 68 | \n",
" 70 | \n",
"
\n",
" \n",
" 2 | \n",
" 75 | \n",
" 72 | \n",
"
\n",
" \n",
" 3 | \n",
" 76 | \n",
" 75 | \n",
"
\n",
" \n",
" 4 | \n",
" 79 | \n",
" 78 | \n",
"
\n",
" \n",
" 5 | \n",
" 73 | \n",
" 78 | \n",
"
\n",
" \n",
" 6 | \n",
" 79 | \n",
" 80 | \n",
"
\n",
" \n",
" 7 | \n",
" 85 | \n",
" 83 | \n",
"
\n",
" \n",
" 8 | \n",
" 89 | \n",
" 87 | \n",
"
\n",
" \n",
" 9 | \n",
" 87 | \n",
" 87 | \n",
"
\n",
" \n",
" 10 | \n",
" 92 | \n",
" 90 | \n",
"
\n",
" \n",
" 11 | \n",
" 93 | \n",
" 90 | \n",
"
\n",
" \n",
" 12 | \n",
" 94 | \n",
" 92 | \n",
"
\n",
" \n",
" 13 | \n",
" 91 | \n",
" 93 | \n",
"
\n",
" \n",
" 14 | \n",
" 97 | \n",
" 95 | \n",
"
\n",
" \n",
" 15 | \n",
" 98 | \n",
" 100 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" final midterm\n",
"0 72 65\n",
"1 68 70\n",
"2 75 72\n",
"3 76 75\n",
"4 79 78\n",
"5 73 78\n",
"6 79 80\n",
"7 85 83\n",
"8 89 87\n",
"9 87 87\n",
"10 92 90\n",
"11 93 90\n",
"12 94 92\n",
"13 91 93\n",
"14 97 95\n",
"15 98 100"
]
},
"execution_count": 9,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"midterm = [65, 70, 72, 75, 78, 78, 80, 83, 87, 87, 90, 90, 92, 93, 95, 100]\n",
"final = [72, 68, 75, 76, 79, 73, 79, 85, 89, 87, 92, 93, 94, 91, 97, 98]\n",
"grades = DataFrame({'midterm': midterm, 'final': final})\n",
"grades"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Task 3: Can you create that scatter plot?
\n",
"\n",
"\n"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"# Your code here"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## The Medicine Hat Tigers\n",
"\n",
"\n",
"\n",
"Let's load in some data about players from the Medicine Hat Tigers.\n",
"After that we'll create a few new columns:\n",
"\n",
"* `heightcm` is the height of the player in centimeters (note that the Height column is the player's height in inches)\n",
"* `hometownLength` is the number of characters in the string representing the hometown. for example, number 11, Jordon Hickmott's hometown is Mission BC which has 10 letters as indicated in hometownLength."
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" Name | \n",
" Pos. | \n",
" L/R | \n",
" Height | \n",
" Weight | \n",
" Birthday | \n",
" Hometown | \n",
" heightcm | \n",
" hometownLength | \n",
"
\n",
" \n",
" \n",
" \n",
" 0 | \n",
" Brennan Bosch | \n",
" C | \n",
" R | \n",
" 68 | \n",
" 173 | \n",
" 02/14/88 | \n",
" Martensville, SK | \n",
" 172.72 | \n",
" 16 | \n",
"
\n",
" \n",
" 1 | \n",
" Scott Wasden | \n",
" C | \n",
" R | \n",
" 73 | \n",
" 188 | \n",
" 01/04/88 | \n",
" Westbank, BC | \n",
" 185.42 | \n",
" 12 | \n",
"
\n",
" \n",
" 2 | \n",
" Colton Grant | \n",
" LW | \n",
" L | \n",
" 69 | \n",
" 177 | \n",
" 03/20/89 | \n",
" Standard, AB | \n",
" 175.26 | \n",
" 12 | \n",
"
\n",
" \n",
" 3 | \n",
" Darren Helm | \n",
" LW | \n",
" L | \n",
" 72 | \n",
" 182 | \n",
" 01/21/87 | \n",
" St. Andrews, MB | \n",
" 182.88 | \n",
" 15 | \n",
"
\n",
" \n",
" 4 | \n",
" Derek Dorsett | \n",
" RW | \n",
" L | \n",
" 71 | \n",
" 178 | \n",
" 12/20/86 | \n",
" Kindersley, SK | \n",
" 180.34 | \n",
" 14 | \n",
"
\n",
" \n",
" 5 | \n",
" Daine Todd | \n",
" C | \n",
" R | \n",
" 70 | \n",
" 173 | \n",
" 01/10/87 | \n",
" Red Deer, AB | \n",
" 177.80 | \n",
" 12 | \n",
"
\n",
" \n",
" 6 | \n",
" Tyler Swystun | \n",
" RW | \n",
" R | \n",
" 71 | \n",
" 185 | \n",
" 01/15/88 | \n",
" Cochrane, AB | \n",
" 180.34 | \n",
" 12 | \n",
"
\n",
" \n",
" 7 | \n",
" Matt Lowry | \n",
" C | \n",
" R | \n",
" 72 | \n",
" 186 | \n",
" 03/02/88 | \n",
" Neepawa, MB | \n",
" 182.88 | \n",
" 11 | \n",
"
\n",
" \n",
" 8 | \n",
" Kevin Undershute | \n",
" LW | \n",
" L | \n",
" 72 | \n",
" 178 | \n",
" 04/12/87 | \n",
" Medicine Hat, AB | \n",
" 182.88 | \n",
" 16 | \n",
"
\n",
" \n",
" 9 | \n",
" Jerrid Sauer | \n",
" RW | \n",
" R | \n",
" 71 | \n",
" 196 | \n",
" 09/12/87 | \n",
" Medicine Hat, AB | \n",
" 180.34 | \n",
" 16 | \n",
"
\n",
" \n",
" 10 | \n",
" Tyler Ennis | \n",
" C | \n",
" L | \n",
" 69 | \n",
" 160 | \n",
" 10/06/89 | \n",
" Edmonton, AB | \n",
" 175.26 | \n",
" 12 | \n",
"
\n",
" \n",
" 11 | \n",
" Jordan Hickmott | \n",
" C | \n",
" R | \n",
" 72 | \n",
" 183 | \n",
" 04/11/90 | \n",
" Mission BC | \n",
" 182.88 | \n",
" 10 | \n",
"
\n",
" \n",
" 12 | \n",
" Jakub Rumpel | \n",
" RW | \n",
" R | \n",
" 68 | \n",
" 166 | \n",
" 01/27/87 | \n",
" Hrnciarovce, SLO | \n",
" 172.72 | \n",
" 16 | \n",
"
\n",
" \n",
" 13 | \n",
" Bretton Cameron | \n",
" C | \n",
" R | \n",
" 71 | \n",
" 168 | \n",
" 01/26/89 | \n",
" Didsbury, AB | \n",
" 180.34 | \n",
" 12 | \n",
"
\n",
" \n",
" 14 | \n",
" Chris Stevens | \n",
" LW | \n",
" L | \n",
" 70 | \n",
" 197 | \n",
" 08/20/86 | \n",
" Dawson Creek, BC | \n",
" 177.80 | \n",
" 16 | \n",
"
\n",
" \n",
" 15 | \n",
" Gord Baldwin | \n",
" D | \n",
" L | \n",
" 77 | \n",
" 205 | \n",
" 03/01/87 | \n",
" Winnipeg, MB | \n",
" 195.58 | \n",
" 12 | \n",
"
\n",
" \n",
" 16 | \n",
" David Schlemko | \n",
" D | \n",
" L | \n",
" 73 | \n",
" 195 | \n",
" 05/07/87 | \n",
" Edmonton, AB | \n",
" 185.42 | \n",
" 12 | \n",
"
\n",
" \n",
" 17 | \n",
" Trever Glass | \n",
" D | \n",
" L | \n",
" 72 | \n",
" 190 | \n",
" 01/22/88 | \n",
" Cochrane, AB | \n",
" 182.88 | \n",
" 12 | \n",
"
\n",
" \n",
" 18 | \n",
" Kris Russell | \n",
" D | \n",
" L | \n",
" 70 | \n",
" 177 | \n",
" 05/02/87 | \n",
" Caroline, AB | \n",
" 177.80 | \n",
" 12 | \n",
"
\n",
" \n",
" 19 | \n",
" Michael Sauer | \n",
" D | \n",
" R | \n",
" 73 | \n",
" 205 | \n",
" 08/07/87 | \n",
" Sartell, MN | \n",
" 185.42 | \n",
" 11 | \n",
"
\n",
" \n",
" 20 | \n",
" Mark Isherwood | \n",
" D | \n",
" R | \n",
" 72 | \n",
" 183 | \n",
" 01/31/89 | \n",
" Abbotsford, BC | \n",
" 182.88 | \n",
" 14 | \n",
"
\n",
" \n",
" 21 | \n",
" Shayne Brown | \n",
" D | \n",
" L | \n",
" 73 | \n",
" 198 | \n",
" 02/20/89 | \n",
" Stony Plain, AB | \n",
" 185.42 | \n",
" 15 | \n",
"
\n",
" \n",
" 22 | \n",
" Jordan Bendfeld | \n",
" D | \n",
" R | \n",
" 75 | \n",
" 230 | \n",
" 02/09/88 | \n",
" Ledue, AB | \n",
" 190.50 | \n",
" 9 | \n",
"
\n",
" \n",
" 23 | \n",
" Ryan Holfeld | \n",
" G | \n",
" L | \n",
" 71 | \n",
" 166 | \n",
" 06/29/89 | \n",
" LeRoy, SK | \n",
" 180.34 | \n",
" 9 | \n",
"
\n",
" \n",
" 24 | \n",
" Matt Keetley | \n",
" G | \n",
" R | \n",
" 74 | \n",
" 189 | \n",
" 04/27/86 | \n",
" Medicine Hat, AB | \n",
" 187.96 | \n",
" 16 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Name Pos. L/R Height Weight Birthday Hometown \\\n",
"0 Brennan Bosch C R 68 173 02/14/88 Martensville, SK \n",
"1 Scott Wasden C R 73 188 01/04/88 Westbank, BC \n",
"2 Colton Grant LW L 69 177 03/20/89 Standard, AB \n",
"3 Darren Helm LW L 72 182 01/21/87 St. Andrews, MB \n",
"4 Derek Dorsett RW L 71 178 12/20/86 Kindersley, SK \n",
"5 Daine Todd C R 70 173 01/10/87 Red Deer, AB \n",
"6 Tyler Swystun RW R 71 185 01/15/88 Cochrane, AB \n",
"7 Matt Lowry C R 72 186 03/02/88 Neepawa, MB \n",
"8 Kevin Undershute LW L 72 178 04/12/87 Medicine Hat, AB \n",
"9 Jerrid Sauer RW R 71 196 09/12/87 Medicine Hat, AB \n",
"10 Tyler Ennis C L 69 160 10/06/89 Edmonton, AB \n",
"11 Jordan Hickmott C R 72 183 04/11/90 Mission BC \n",
"12 Jakub Rumpel RW R 68 166 01/27/87 Hrnciarovce, SLO \n",
"13 Bretton Cameron C R 71 168 01/26/89 Didsbury, AB \n",
"14 Chris Stevens LW L 70 197 08/20/86 Dawson Creek, BC \n",
"15 Gord Baldwin D L 77 205 03/01/87 Winnipeg, MB \n",
"16 David Schlemko D L 73 195 05/07/87 Edmonton, AB \n",
"17 Trever Glass D L 72 190 01/22/88 Cochrane, AB \n",
"18 Kris Russell D L 70 177 05/02/87 Caroline, AB \n",
"19 Michael Sauer D R 73 205 08/07/87 Sartell, MN \n",
"20 Mark Isherwood D R 72 183 01/31/89 Abbotsford, BC \n",
"21 Shayne Brown D L 73 198 02/20/89 Stony Plain, AB \n",
"22 Jordan Bendfeld D R 75 230 02/09/88 Ledue, AB \n",
"23 Ryan Holfeld G L 71 166 06/29/89 LeRoy, SK \n",
"24 Matt Keetley G R 74 189 04/27/86 Medicine Hat, AB \n",
"\n",
" heightcm hometownLength \n",
"0 172.72 16 \n",
"1 185.42 12 \n",
"2 175.26 12 \n",
"3 182.88 15 \n",
"4 180.34 14 \n",
"5 177.80 12 \n",
"6 180.34 12 \n",
"7 182.88 11 \n",
"8 182.88 16 \n",
"9 180.34 16 \n",
"10 175.26 12 \n",
"11 182.88 10 \n",
"12 172.72 16 \n",
"13 180.34 12 \n",
"14 177.80 16 \n",
"15 195.58 12 \n",
"16 185.42 12 \n",
"17 182.88 12 \n",
"18 177.80 12 \n",
"19 185.42 11 \n",
"20 182.88 14 \n",
"21 185.42 15 \n",
"22 190.50 9 \n",
"23 180.34 9 \n",
"24 187.96 16 "
]
},
"execution_count": 12,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"medicineHat = pd.read_csv('https://raw.githubusercontent.com/zacharski/machine-learning/master/data/medicineHatTigers.csv')\n",
"medicineHat['heightcm'] = medicineHat['Height'] * 2.54\n",
"medicineHat['hometownLength'] = medicineHat['Hometown'].str.len()\n",
"medicineHat\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Let's explore this data a bit. (Some of these are screwy examples but they illustrate a point)\n",
"\n",
"Task 4: Can you create a scatter plot showing the correlation between height in inches and height in cm?
\n",
"\n",
"\n",
"\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": []
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Task 5: A scatter plot showing the correlation between a person's height in inches and the number of characters in the hometown
\n",
"\n",
"\n",
"\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": []
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Task 6: A scatter plot showing the correlation between a person's height in cm and their weight.
\n",
"\n",
"\n",
"\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": []
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": true
},
"source": [
"Task 7: Which of these has the best correlation? Which shows no correlation?
\n",
"As you can see, if the scatter plot is on a straight line it has perfect correlation.\n",
"\n",
"And if dots are scattered all over the place than there is likely no correlation. \n",
"\n",
"\n",
"Let's look at some other lines:"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## negative correlation has a line going downhill\n",
"\n",
"#### the relationship of hair length to a person's height"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAZcAAAEWCAYAAACqitpwAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEFSkmXs5IgAAAABJRU5ErkJggg==\n",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"height =[62, 64, 65, 68, 69, 70, 67, 65, 72, 73, 74]\n",
"hairLength = [7, 10, 6, 4, 5, 4, 5, 8, 1, 1, 3]\n",
"cm = [173, 185, 175, 183, 180, 196]\n",
"people = DataFrame({'height': height, 'hairLength': hairLength})\n",
"\n",
"plt.scatter(people['height'], people['hairLength'])\n",
"\n",
"\n",
"xlab = 'Amount Spent on Makeup [in Yen]'\n",
"ylab = 'Amount Spent on Clothes [in Yen]'\n",
"title = 'Weight and calories in 1 to 3 yr. old children'\n",
"\n",
"# Add axis labels\n",
"plt.xlabel(\"Height in Inches\")\n",
"plt.ylabel(\"Hair Length in Inches\")\n",
"\n",
"\n",
"# Add title\n",
"plt.title(\"Comparison of an Individual's Height to the Length of their Hair\")\n",
"\n",
"\n",
"# Show plot\n",
"plt.show()\n",
"\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"So The taller someone is, the shorter their hair.\n",
"Task 7: A question
\n",
"Do you think correlation means causation? This means that Do you think as a person gets taller somehow that causes a perons's hair to be shorter?\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"## The Effects of Alcohol on the Dexterity Score\n",
"#### a strong negative correlation"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEWCAYAAAB8LwAVAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSUVORK5CYII=\n",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"drinks = list(range(11))\n",
"dexterity = [96, 96, 90, 63, 65, 50, 47, 46, 18, 17, 9]\n",
"drinking = DataFrame({'drinks': drinks, 'dexterity': dexterity})\n",
"\n",
"plt.scatter(drinking['drinks'], drinking['dexterity'])\n",
"\n",
"\n",
"\n",
"# Add axis labels\n",
"plt.xlabel(\"Number of Drinks\")\n",
"plt.ylabel(\"Dexterity Score\")\n",
"\n",
"\n",
"# Add title\n",
"plt.title(\"The effects of alcohol consumption on dexterity\")\n",
"\n",
"\n",
"# Show plot\n",
"plt.show()\n",
"\n",
"\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Numeric Represenstation of the Strength of the Correlation\n",
"\n",
"\n",
"So far we've looked at correlations visually. The more a scatter plot looks like a straight line the stronger the correlation. If the line goes up from left to right it is a positive correlation. If it goes down it is a negative correlation. We can also determine correlation using a number...\n",
"\n",
"## Pearson Correlation Coefficient\n",
"\n",
"The Pearson Correlation Coefficient ranges from -1 to 1.\n",
"1 is perfect positive correlation, -1 is perfect negative. The formula is as follows:\n",
"\n",
"$$r=\\frac{\\sum_{i=1}^n(x_i - \\bar{x})(y_i-\\bar{y})}{\\sqrt{\\sum_{i=1}^n(x_i - \\bar{x})} \\sqrt{\\sum_{i=1}^n(y_i - \\bar{y})}}$$\n",
"\n",
"Let me say immediately that you don't need to know that formula for this class. Should you dive deeper into Data Science you will need to interpret and apply formulas like this, but the good news is that these formulas always look a lot harder than they actually are. (Take it from me -- I had zero real math classes in high school and none in college (I have a Bachelor of Fine Arts degree))\n",
"\n",
"Anyway, let's ignore the formula and think that a number near 1 means a positive correlation; 0 means no correlation, and near -1 means a strong negative correlation. \n",
"\n",
"### Japanese ladies expensives on makeup and clothes\n",
"Let's go back to our first example.\n",
"First here is the data:"
]
},
{
"cell_type": "code",
"execution_count": 77,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"
\n",
" \n",
" \n",
" | \n",
" clothes | \n",
" makeup | \n",
"
\n",
" \n",
" \n",
" \n",
" Ms A | \n",
" 7000 | \n",
" 3000 | \n",
"
\n",
" \n",
" Ms B | \n",
" 8000 | \n",
" 5000 | \n",
"
\n",
" \n",
" Ms C | \n",
" 25000 | \n",
" 12000 | \n",
"
\n",
" \n",
" Ms D | \n",
" 5000 | \n",
" 2000 | \n",
"
\n",
" \n",
" Ms E | \n",
" 12000 | \n",
" 7000 | \n",
"
\n",
" \n",
" Ms F | \n",
" 30000 | \n",
" 15000 | \n",
"
\n",
" \n",
" Ms G | \n",
" 10000 | \n",
" 5000 | \n",
"
\n",
" \n",
" Ms H | \n",
" 15000 | \n",
" 6000 | \n",
"
\n",
" \n",
" Ms I | \n",
" 20000 | \n",
" 8000 | \n",
"
\n",
" \n",
" Ms J | \n",
" 18000 | \n",
" 10000 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" clothes makeup\n",
"Ms A 7000 3000\n",
"Ms B 8000 5000\n",
"Ms C 25000 12000\n",
"Ms D 5000 2000\n",
"Ms E 12000 7000\n",
"Ms F 30000 15000\n",
"Ms G 10000 5000\n",
"Ms H 15000 6000\n",
"Ms I 20000 8000\n",
"Ms J 18000 10000"
]
},
"execution_count": 77,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"monthly"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"And our Scatter plot looked like this:"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAYcAAAD8CAYAAACcjGjIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Import package\n",
"import matplotlib.pyplot as plt\n",
"\n",
"# Build Scatter plot\n",
"plt.scatter(monthly['makeup'], monthly['clothes'])\n",
"\n",
"\n",
"# Show plot\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"So that looks like a pretty strong positive correlation. To compute Pearson on this data we do:"
]
},
{
"cell_type": "code",
"execution_count": 80,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"
\n",
" \n",
" \n",
" | \n",
" clothes | \n",
" makeup | \n",
"
\n",
" \n",
" \n",
" \n",
" clothes | \n",
" 1.00000 | \n",
" 0.96802 | \n",
"
\n",
" \n",
" makeup | \n",
" 0.96802 | \n",
" 1.00000 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" clothes makeup\n",
"clothes 1.00000 0.96802\n",
"makeup 0.96802 1.00000"
]
},
"execution_count": 80,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"monthly.corr()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"So we have a nice table with 1s on the diagonal. The diagonal states the obvious: clothes expenditures are perfectly correlated with clothes expenditures and makeup expenditures are perfectly correlated with makeup expenditures. \n",
"\n",
"The other cells are the interesting ones and they show that the correlation between clothes and makeup is very strong at 0.96802.\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Let's look at the correlation between the number of drinks and dexterity: "
]
},
{
"cell_type": "code",
"execution_count": 83,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"
\n",
" \n",
" \n",
" | \n",
" dexterity | \n",
" drinks | \n",
"
\n",
" \n",
" \n",
" \n",
" dexterity | \n",
" 1.000000 | \n",
" -0.979237 | \n",
"
\n",
" \n",
" drinks | \n",
" -0.979237 | \n",
" 1.000000 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" dexterity drinks\n",
"dexterity 1.000000 -0.979237\n",
"drinks -0.979237 1.000000"
]
},
"execution_count": 83,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"drinking.corr()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"So -0.97923 indicates a very strong negative correlation"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": true
},
"source": [
"Task 8: School Music Program: a case study
\n",
"\n",
"We are interested in seeing whether a relationship exists between music performance\n",
"grades and students' individual Iowa Test of Basic Skills (ITBS) scores. Let's load in the data"
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" Band | \n",
" Math | \n",
" Language | \n",
" Science | \n",
"
\n",
" \n",
" Student | \n",
" | \n",
" | \n",
" | \n",
" | \n",
"
\n",
" \n",
" \n",
" \n",
" 1 | \n",
" 3 | \n",
" 220 | \n",
" 215 | \n",
" 220 | \n",
"
\n",
" \n",
" 2 | \n",
" 4 | \n",
" 240 | \n",
" 210 | \n",
" 225 | \n",
"
\n",
" \n",
" 3 | \n",
" 3 | \n",
" 210 | \n",
" 250 | \n",
" 235 | \n",
"
\n",
" \n",
" 4 | \n",
" 2 | \n",
" 215 | \n",
" 230 | \n",
" 210 | \n",
"
\n",
" \n",
" 5 | \n",
" 4 | \n",
" 260 | \n",
" 240 | \n",
" 220 | \n",
"
\n",
" \n",
" 6 | \n",
" 3 | \n",
" 230 | \n",
" 270 | \n",
" 250 | \n",
"
\n",
" \n",
" 7 | \n",
" 4 | \n",
" 240 | \n",
" 240 | \n",
" 220 | \n",
"
\n",
" \n",
" 8 | \n",
" 4 | \n",
" 259 | \n",
" 220 | \n",
" 240 | \n",
"
\n",
" \n",
" 9 | \n",
" 3 | \n",
" 245 | \n",
" 230 | \n",
" 250 | \n",
"
\n",
" \n",
" 10 | \n",
" 4 | \n",
" 280 | \n",
" 270 | \n",
" 230 | \n",
"
\n",
" \n",
" 11 | \n",
" 5 | \n",
" 300 | \n",
" 260 | \n",
" 220 | \n",
"
\n",
" \n",
" 12 | \n",
" 2 | \n",
" 230 | \n",
" 250 | \n",
" 225 | \n",
"
\n",
" \n",
" 13 | \n",
" 3 | \n",
" 250 | \n",
" 245 | \n",
" 235 | \n",
"
\n",
" \n",
" 14 | \n",
" 4 | \n",
" 275 | \n",
" 235 | \n",
" 210 | \n",
"
\n",
" \n",
" 15 | \n",
" 3 | \n",
" 200 | \n",
" 260 | \n",
" 220 | \n",
"
\n",
" \n",
" 16 | \n",
" 2 | \n",
" 200 | \n",
" 255 | \n",
" 250 | \n",
"
\n",
" \n",
" 17 | \n",
" 4 | \n",
" 290 | \n",
" 250 | \n",
" 220 | \n",
"
\n",
" \n",
" 18 | \n",
" 3 | \n",
" 250 | \n",
" 230 | \n",
" 240 | \n",
"
\n",
" \n",
" 19 | \n",
" 4 | \n",
" 270 | \n",
" 245 | \n",
" 250 | \n",
"
\n",
" \n",
" 20 | \n",
" 4 | \n",
" 260 | \n",
" 270 | \n",
" 230 | \n",
"
\n",
" \n",
" 21 | \n",
" 3 | \n",
" 260 | \n",
" 270 | \n",
" 260 | \n",
"
\n",
" \n",
" 22 | \n",
" 4 | \n",
" 270 | \n",
" 250 | \n",
" 230 | \n",
"
\n",
" \n",
" 23 | \n",
" 5 | \n",
" 310 | \n",
" 260 | \n",
" 260 | \n",
"
\n",
" \n",
" 24 | \n",
" 2 | \n",
" 170 | \n",
" 250 | \n",
" 220 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Band Math Language Science\n",
"Student \n",
"1 3 220 215 220\n",
"2 4 240 210 225\n",
"3 3 210 250 235\n",
"4 2 215 230 210\n",
"5 4 260 240 220\n",
"6 3 230 270 250\n",
"7 4 240 240 220\n",
"8 4 259 220 240\n",
"9 3 245 230 250\n",
"10 4 280 270 230\n",
"11 5 300 260 220\n",
"12 2 230 250 225\n",
"13 3 250 245 235\n",
"14 4 275 235 210\n",
"15 3 200 260 220\n",
"16 2 200 255 250\n",
"17 4 290 250 220\n",
"18 3 250 230 240\n",
"19 4 270 245 250\n",
"20 4 260 270 230\n",
"21 3 260 270 260\n",
"22 4 270 250 230\n",
"23 5 310 260 260\n",
"24 2 170 250 220"
]
},
"execution_count": 25,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"musicFile = \"https://raw.githubusercontent.com/zacharski/machine-learning/master/data/musicITBS.csv\"\n",
"music = pd.read_csv(musicFile)\n",
"music.set_index('Student', inplace=True)\n",
"\n",
"music"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We want to know what test score (math, language, science) that the music performance grade is most highly correlated with. Write code that computes the correlation table. Also, for the most highly correlated pair, please draw a scatter plot. For example, if science is most highly correlated with band draw that scatter plot.\n"
]
},
{
"cell_type": "code",
"execution_count": 29,
"metadata": {},
"outputs": [],
"source": [
"# Your code here\n",
"\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Using a scatter plot graph the strongest association\n",
"\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
" "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Task 9: What do you think?
\n",
"\n",
"1. Is knowing a person's music performance grade a good predictor for that person's score in math? language? science?\n",
"\n",
"2. Does this study offer support for an argument not to eliminate the music program at our high school because it improves math scores?\n",
"\n",
"3. Does this study offer support for an argument that eliminating the music program will have no effect on language scores?"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": []
}
],
"metadata": {
"anaconda-cloud": {},
"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": "3.6.1"
}
},
"nbformat": 4,
"nbformat_minor": 1
}