A simple simulation model of job market

jobmarket.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "using DataFrames, CSV, Statistics, PyPlot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "┌ Info: Fetching data from a remote repository\n",
      "└ @ Main In[2]:4\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<table class=\"data-frame\"><thead><tr><th></th><th>Column1</th><th>stateur</th><th>statemb</th><th>state</th><th>age</th><th>tenure</th><th>joblost</th><th>nwhite</th><th>school12</th><th>sex</th><th>bluecol</th><th>smsa</th><th>married</th><th>dkids</th><th>dykids</th><th>yrdispl</th><th>rr</th><th>head</th><th>ui</th></tr><tr><th></th><th>Int64⍰</th><th>Float64⍰</th><th>Int64⍰</th><th>Int64⍰</th><th>Int64⍰</th><th>Int64⍰</th><th>String⍰</th><th>String⍰</th><th>String⍰</th><th>String⍰</th><th>String⍰</th><th>String⍰</th><th>String⍰</th><th>String⍰</th><th>String⍰</th><th>Int64⍰</th><th>Float64⍰</th><th>String⍰</th><th>String⍰</th></tr></thead><tbody><p>4,877 rows × 19 columns</p><tr><th>1</th><td>1</td><td>4.5</td><td>167</td><td>42</td><td>49</td><td>21</td><td>other</td><td>no</td><td>no</td><td>male</td><td>yes</td><td>yes</td><td>no</td><td>no</td><td>no</td><td>7</td><td>0.290631</td><td>yes</td><td>yes</td></tr><tr><th>2</th><td>2</td><td>10.5</td><td>251</td><td>55</td><td>26</td><td>2</td><td>slack_work</td><td>no</td><td>no</td><td>male</td><td>yes</td><td>yes</td><td>no</td><td>yes</td><td>yes</td><td>10</td><td>0.520202</td><td>yes</td><td>no</td></tr><tr><th>3</th><td>3</td><td>7.2</td><td>260</td><td>21</td><td>40</td><td>19</td><td>other</td><td>no</td><td>yes</td><td>female</td><td>yes</td><td>yes</td><td>yes</td><td>no</td><td>no</td><td>10</td><td>0.43249</td><td>yes</td><td>yes</td></tr><tr><th>4</th><td>4</td><td>5.8</td><td>245</td><td>56</td><td>51</td><td>17</td><td>slack_work</td><td>yes</td><td>no</td><td>female</td><td>yes</td><td>yes</td><td>yes</td><td>no</td><td>no</td><td>10</td><td>0.5</td><td>no</td><td>yes</td></tr><tr><th>5</th><td>5</td><td>6.5</td><td>125</td><td>58</td><td>33</td><td>1</td><td>slack_work</td><td>no</td><td>yes</td><td>male</td><td>yes</td><td>yes</td><td>yes</td><td>yes</td><td>yes</td><td>4</td><td>0.390625</td><td>yes</td><td>no</td></tr><tr><th>6</th><td>6</td><td>7.5</td><td>188</td><td>11</td><td>51</td><td>3</td><td>other</td><td>no</td><td>no</td><td>male</td><td>yes</td><td>yes</td><td>yes</td><td>no</td><td>no</td><td>10</td><td>0.482201</td><td>yes</td><td>yes</td></tr><tr><th>7</th><td>7</td><td>5.8</td><td>166</td><td>93</td><td>30</td><td>5</td><td>position_abolished</td><td>no</td><td>no</td><td>male</td><td>yes</td><td>yes</td><td>yes</td><td>yes</td><td>yes</td><td>6</td><td>0.334043</td><td>yes</td><td>yes</td></tr><tr><th>8</th><td>8</td><td>5.8</td><td>214</td><td>84</td><td>26</td><td>3</td><td>slack_work</td><td>no</td><td>no</td><td>male</td><td>yes</td><td>yes</td><td>yes</td><td>yes</td><td>yes</td><td>8</td><td>0.510989</td><td>yes</td><td>yes</td></tr><tr><th>9</th><td>9</td><td>7.7</td><td>213</td><td>84</td><td>54</td><td>20</td><td>other</td><td>no</td><td>no</td><td>male</td><td>yes</td><td>yes</td><td>yes</td><td>no</td><td>no</td><td>6</td><td>0.355634</td><td>yes</td><td>no</td></tr><tr><th>10</th><td>10</td><td>6.0</td><td>187</td><td>33</td><td>31</td><td>1</td><td>slack_work</td><td>no</td><td>no</td><td>male</td><td>yes</td><td>yes</td><td>yes</td><td>no</td><td>no</td><td>8</td><td>0.373272</td><td>yes</td><td>yes</td></tr><tr><th>11</th><td>11</td><td>7.1</td><td>206</td><td>33</td><td>56</td><td>20</td><td>slack_work</td><td>no</td><td>no</td><td>male</td><td>yes</td><td>no</td><td>yes</td><td>no</td><td>no</td><td>10</td><td>0.404467</td><td>yes</td><td>yes</td></tr><tr><th>12</th><td>12</td><td>10.1</td><td>132</td><td>12</td><td>54</td><td>26</td><td>other</td><td>no</td><td>no</td><td>male</td><td>yes</td><td>no</td><td>yes</td><td>yes</td><td>no</td><td>2</td><td>0.413295</td><td>yes</td><td>no</td></tr><tr><th>13</th><td>13</td><td>5.0</td><td>279</td><td>22</td><td>21</td><td>1</td><td>slack_work</td><td>no</td><td>no</td><td>male</td><td>yes</td><td>yes</td><td>no</td><td>no</td><td>no</td><td>9</td><td>0.598485</td><td>no</td><td>no</td></tr><tr><th>14</th><td>14</td><td>6.6</td><td>291</td><td>22</td><td>35</td><td>1</td><td>slack_work</td><td>yes</td><td>no</td><td>male</td><td>yes</td><td>yes</td><td>no</td><td>yes</td><td>yes</td><td>10</td><td>0.599156</td><td>no</td><td>no</td></tr><tr><th>15</th><td>15</td><td>7.2</td><td>166</td><td>93</td><td>23</td><td>1</td><td>slack_work</td><td>no</td><td>yes</td><td>male</td><td>yes</td><td>yes</td><td>no</td><td>no</td><td>no</td><td>4</td><td>0.39375</td><td>no</td><td>yes</td></tr><tr><th>16</th><td>16</td><td>7.2</td><td>260</td><td>21</td><td>41</td><td>3</td><td>other</td><td>no</td><td>no</td><td>male</td><td>yes</td><td>yes</td><td>yes</td><td>yes</td><td>no</td><td>10</td><td>0.502304</td><td>no</td><td>yes</td></tr><tr><th>17</th><td>17</td><td>6.4</td><td>213</td><td>84</td><td>49</td><td>12</td><td>other</td><td>no</td><td>no</td><td>male</td><td>yes</td><td>yes</td><td>yes</td><td>no</td><td>no</td><td>7</td><td>0.426374</td><td>no</td><td>yes</td></tr><tr><th>18</th><td>18</td><td>8.9</td><td>166</td><td>56</td><td>35</td><td>1</td><td>other</td><td>yes</td><td>no</td><td>male</td><td>yes</td><td>no</td><td>yes</td><td>no</td><td>no</td><td>2</td><td>0.473545</td><td>yes</td><td>no</td></tr><tr><th>19</th><td>19</td><td>5.8</td><td>245</td><td>56</td><td>40</td><td>12</td><td>other</td><td>no</td><td>no</td><td>female</td><td>yes</td><td>yes</td><td>yes</td><td>yes</td><td>no</td><td>10</td><td>0.498195</td><td>no</td><td>yes</td></tr><tr><th>20</th><td>20</td><td>6.1</td><td>138</td><td>54</td><td>60</td><td>30</td><td>slack_work</td><td>no</td><td>no</td><td>male</td><td>yes</td><td>yes</td><td>yes</td><td>no</td><td>no</td><td>2</td><td>0.518518</td><td>yes</td><td>no</td></tr><tr><th>21</th><td>21</td><td>7.2</td><td>260</td><td>21</td><td>42</td><td>4</td><td>slack_work</td><td>yes</td><td>no</td><td>female</td><td>yes</td><td>yes</td><td>yes</td><td>no</td><td>no</td><td>10</td><td>0.5</td><td>no</td><td>yes</td></tr><tr><th>22</th><td>22</td><td>18.0</td><td>211</td><td>55</td><td>48</td><td>3</td><td>other</td><td>no</td><td>no</td><td>male</td><td>yes</td><td>no</td><td>yes</td><td>no</td><td>no</td><td>2</td><td>0.255892</td><td>yes</td><td>yes</td></tr><tr><th>23</th><td>23</td><td>4.6</td><td>168</td><td>16</td><td>41</td><td>3</td><td>other</td><td>no</td><td>no</td><td>male</td><td>yes</td><td>yes</td><td>yes</td><td>no</td><td>no</td><td>3</td><td>0.335907</td><td>yes</td><td>yes</td></tr><tr><th>24</th><td>24</td><td>9.0</td><td>192</td><td>83</td><td>29</td><td>5</td><td>position_abolished</td><td>no</td><td>no</td><td>male</td><td>yes</td><td>no</td><td>yes</td><td>yes</td><td>yes</td><td>5</td><td>0.262204</td><td>yes</td><td>no</td></tr><tr><th>25</th><td>25</td><td>5.9</td><td>185</td><td>44</td><td>59</td><td>5</td><td>slack_work</td><td>no</td><td>no</td><td>male</td><td>yes</td><td>no</td><td>yes</td><td>no</td><td>no</td><td>4</td><td>0.5</td><td>yes</td><td>yes</td></tr><tr><th>26</th><td>26</td><td>7.4</td><td>197</td><td>73</td><td>32</td><td>3</td><td>slack_work</td><td>no</td><td>no</td><td>male</td><td>yes</td><td>no</td><td>yes</td><td>yes</td><td>yes</td><td>6</td><td>0.438967</td><td>yes</td><td>yes</td></tr><tr><th>27</th><td>27</td><td>11.4</td><td>168</td><td>33</td><td>22</td><td>3</td><td>slack_work</td><td>no</td><td>no</td><td>male</td><td>yes</td><td>yes</td><td>no</td><td>yes</td><td>no</td><td>2</td><td>0.504065</td><td>no</td><td>no</td></tr><tr><th>28</th><td>28</td><td>5.5</td><td>150</td><td>43</td><td>25</td><td>3</td><td>slack_work</td><td>yes</td><td>no</td><td>male</td><td>yes</td><td>yes</td><td>yes</td><td>yes</td><td>yes</td><td>8</td><td>0.575221</td><td>no</td><td>no</td></tr><tr><th>29</th><td>29</td><td>4.8</td><td>179</td><td>44</td><td>36</td><td>3</td><td>other</td><td>no</td><td>yes</td><td>male</td><td>yes</td><td>no</td><td>yes</td><td>yes</td><td>yes</td><td>7</td><td>0.401099</td><td>yes</td><td>yes</td></tr><tr><th>30</th><td>30</td><td>6.3</td><td>150</td><td>59</td><td>41</td><td>20</td><td>other</td><td>no</td><td>no</td><td>male</td><td>yes</td><td>yes</td><td>no</td><td>yes</td><td>no</td><td>3</td><td>0.399485</td><td>no</td><td>yes</td></tr><tr><th>&vellip;</th><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td></tr></tbody></table>"
      ],
      "text/latex": [
       "\\begin{tabular}{r|ccccccccccccccccccc}\n",
       "\t& Column1 & stateur & statemb & state & age & tenure & joblost & nwhite & school12 & sex & bluecol & smsa & married & dkids & dykids & yrdispl & rr & head & ui\\\\\n",
       "\t\\hline\n",
       "\t& Int64⍰ & Float64⍰ & Int64⍰ & Int64⍰ & Int64⍰ & Int64⍰ & String⍰ & String⍰ & String⍰ & String⍰ & String⍰ & String⍰ & String⍰ & String⍰ & String⍰ & Int64⍰ & Float64⍰ & String⍰ & String⍰\\\\\n",
       "\t\\hline\n",
       "\t1 & 1 & 4.5 & 167 & 42 & 49 & 21 & other & no & no & male & yes & yes & no & no & no & 7 & 0.290631 & yes & yes \\\\\n",
       "\t2 & 2 & 10.5 & 251 & 55 & 26 & 2 & slack\\_work & no & no & male & yes & yes & no & yes & yes & 10 & 0.520202 & yes & no \\\\\n",
       "\t3 & 3 & 7.2 & 260 & 21 & 40 & 19 & other & no & yes & female & yes & yes & yes & no & no & 10 & 0.43249 & yes & yes \\\\\n",
       "\t4 & 4 & 5.8 & 245 & 56 & 51 & 17 & slack\\_work & yes & no & female & yes & yes & yes & no & no & 10 & 0.5 & no & yes \\\\\n",
       "\t5 & 5 & 6.5 & 125 & 58 & 33 & 1 & slack\\_work & no & yes & male & yes & yes & yes & yes & yes & 4 & 0.390625 & yes & no \\\\\n",
       "\t6 & 6 & 7.5 & 188 & 11 & 51 & 3 & other & no & no & male & yes & yes & yes & no & no & 10 & 0.482201 & yes & yes \\\\\n",
       "\t7 & 7 & 5.8 & 166 & 93 & 30 & 5 & position\\_abolished & no & no & male & yes & yes & yes & yes & yes & 6 & 0.334043 & yes & yes \\\\\n",
       "\t8 & 8 & 5.8 & 214 & 84 & 26 & 3 & slack\\_work & no & no & male & yes & yes & yes & yes & yes & 8 & 0.510989 & yes & yes \\\\\n",
       "\t9 & 9 & 7.7 & 213 & 84 & 54 & 20 & other & no & no & male & yes & yes & yes & no & no & 6 & 0.355634 & yes & no \\\\\n",
       "\t10 & 10 & 6.0 & 187 & 33 & 31 & 1 & slack\\_work & no & no & male & yes & yes & yes & no & no & 8 & 0.373272 & yes & yes \\\\\n",
       "\t11 & 11 & 7.1 & 206 & 33 & 56 & 20 & slack\\_work & no & no & male & yes & no & yes & no & no & 10 & 0.404467 & yes & yes \\\\\n",
       "\t12 & 12 & 10.1 & 132 & 12 & 54 & 26 & other & no & no & male & yes & no & yes & yes & no & 2 & 0.413295 & yes & no \\\\\n",
       "\t13 & 13 & 5.0 & 279 & 22 & 21 & 1 & slack\\_work & no & no & male & yes & yes & no & no & no & 9 & 0.598485 & no & no \\\\\n",
       "\t14 & 14 & 6.6 & 291 & 22 & 35 & 1 & slack\\_work & yes & no & male & yes & yes & no & yes & yes & 10 & 0.599156 & no & no \\\\\n",
       "\t15 & 15 & 7.2 & 166 & 93 & 23 & 1 & slack\\_work & no & yes & male & yes & yes & no & no & no & 4 & 0.39375 & no & yes \\\\\n",
       "\t16 & 16 & 7.2 & 260 & 21 & 41 & 3 & other & no & no & male & yes & yes & yes & yes & no & 10 & 0.502304 & no & yes \\\\\n",
       "\t17 & 17 & 6.4 & 213 & 84 & 49 & 12 & other & no & no & male & yes & yes & yes & no & no & 7 & 0.426374 & no & yes \\\\\n",
       "\t18 & 18 & 8.9 & 166 & 56 & 35 & 1 & other & yes & no & male & yes & no & yes & no & no & 2 & 0.473545 & yes & no \\\\\n",
       "\t19 & 19 & 5.8 & 245 & 56 & 40 & 12 & other & no & no & female & yes & yes & yes & yes & no & 10 & 0.498195 & no & yes \\\\\n",
       "\t20 & 20 & 6.1 & 138 & 54 & 60 & 30 & slack\\_work & no & no & male & yes & yes & yes & no & no & 2 & 0.518518 & yes & no \\\\\n",
       "\t21 & 21 & 7.2 & 260 & 21 & 42 & 4 & slack\\_work & yes & no & female & yes & yes & yes & no & no & 10 & 0.5 & no & yes \\\\\n",
       "\t22 & 22 & 18.0 & 211 & 55 & 48 & 3 & other & no & no & male & yes & no & yes & no & no & 2 & 0.255892 & yes & yes \\\\\n",
       "\t23 & 23 & 4.6 & 168 & 16 & 41 & 3 & other & no & no & male & yes & yes & yes & no & no & 3 & 0.335907 & yes & yes \\\\\n",
       "\t24 & 24 & 9.0 & 192 & 83 & 29 & 5 & position\\_abolished & no & no & male & yes & no & yes & yes & yes & 5 & 0.262204 & yes & no \\\\\n",
       "\t25 & 25 & 5.9 & 185 & 44 & 59 & 5 & slack\\_work & no & no & male & yes & no & yes & no & no & 4 & 0.5 & yes & yes \\\\\n",
       "\t26 & 26 & 7.4 & 197 & 73 & 32 & 3 & slack\\_work & no & no & male & yes & no & yes & yes & yes & 6 & 0.438967 & yes & yes \\\\\n",
       "\t27 & 27 & 11.4 & 168 & 33 & 22 & 3 & slack\\_work & no & no & male & yes & yes & no & yes & no & 2 & 0.504065 & no & no \\\\\n",
       "\t28 & 28 & 5.5 & 150 & 43 & 25 & 3 & slack\\_work & yes & no & male & yes & yes & yes & yes & yes & 8 & 0.575221 & no & no \\\\\n",
       "\t29 & 29 & 4.8 & 179 & 44 & 36 & 3 & other & no & yes & male & yes & no & yes & yes & yes & 7 & 0.401099 & yes & yes \\\\\n",
       "\t30 & 30 & 6.3 & 150 & 59 & 41 & 20 & other & no & no & male & yes & yes & no & yes & no & 3 & 0.399485 & no & yes \\\\\n",
       "\t$\\dots$ & $\\dots$ & $\\dots$ & $\\dots$ & $\\dots$ & $\\dots$ & $\\dots$ & $\\dots$ & $\\dots$ & $\\dots$ & $\\dots$ & $\\dots$ & $\\dots$ & $\\dots$ & $\\dots$ & $\\dots$ & $\\dots$ & $\\dots$ & $\\dots$ & $\\dots$ \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "4877×19 DataFrame. Omitted printing of 13 columns\n",
       "│ Row  │ Column1 │ stateur  │ statemb │ state  │ age    │ tenure │\n",
       "│      │ \u001b[90mInt64⍰\u001b[39m  │ \u001b[90mFloat64⍰\u001b[39m │ \u001b[90mInt64⍰\u001b[39m  │ \u001b[90mInt64⍰\u001b[39m │ \u001b[90mInt64⍰\u001b[39m │ \u001b[90mInt64⍰\u001b[39m │\n",
       "├──────┼─────────┼──────────┼─────────┼────────┼────────┼────────┤\n",
       "│ 1    │ 1       │ 4.5      │ 167     │ 42     │ 49     │ 21     │\n",
       "│ 2    │ 2       │ 10.5     │ 251     │ 55     │ 26     │ 2      │\n",
       "│ 3    │ 3       │ 7.2      │ 260     │ 21     │ 40     │ 19     │\n",
       "│ 4    │ 4       │ 5.8      │ 245     │ 56     │ 51     │ 17     │\n",
       "│ 5    │ 5       │ 6.5      │ 125     │ 58     │ 33     │ 1      │\n",
       "│ 6    │ 6       │ 7.5      │ 188     │ 11     │ 51     │ 3      │\n",
       "│ 7    │ 7       │ 5.8      │ 166     │ 93     │ 30     │ 5      │\n",
       "│ 8    │ 8       │ 5.8      │ 214     │ 84     │ 26     │ 3      │\n",
       "│ 9    │ 9       │ 7.7      │ 213     │ 84     │ 54     │ 20     │\n",
       "│ 10   │ 10      │ 6.0      │ 187     │ 33     │ 31     │ 1      │\n",
       "⋮\n",
       "│ 4867 │ 4867    │ 4.0      │ 255     │ 14     │ 28     │ 2      │\n",
       "│ 4868 │ 4868    │ 10.5     │ 251     │ 55     │ 32     │ 11     │\n",
       "│ 4869 │ 4869    │ 13.0     │ 225     │ 55     │ 38     │ 12     │\n",
       "│ 4870 │ 4870    │ 9.4      │ 136     │ 93     │ 28     │ 1      │\n",
       "│ 4871 │ 4871    │ 4.0      │ 250     │ 41     │ 25     │ 6      │\n",
       "│ 4872 │ 4872    │ 8.6      │ 125     │ 59     │ 21     │ 2      │\n",
       "│ 4873 │ 4873    │ 7.4      │ 168     │ 33     │ 35     │ 12     │\n",
       "│ 4874 │ 4874    │ 7.0      │ 189     │ 74     │ 39     │ 5      │\n",
       "│ 4875 │ 4875    │ 8.0      │ 168     │ 74     │ 59     │ 13     │\n",
       "│ 4876 │ 4876    │ 6.3      │ 191     │ 41     │ 33     │ 11     │\n",
       "│ 4877 │ 4877    │ 9.3      │ 188     │ 94     │ 27     │ 2      │"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "const FILENAME = \"Benefits.csv\"\n",
    "\n",
    "if !isfile(FILENAME)\n",
    "    @info \"Fetching data from a remote repository\"\n",
    "    download(\"http://vincentarelbundock.github.io/Rdatasets/csv/Ecdat/Benefits.csv\",\n",
    "             FILENAME)\n",
    "end\n",
    "\n",
    "benefits = CSV.read(FILENAME)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.4864526378878213"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cor(benefits.age, benefits.tenure)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"data-frame\"><thead><tr><th></th><th>age</th><th>mean_tenure</th><th>count</th></tr><tr><th></th><th>Int64⍰</th><th>Float64</th><th>Int64</th></tr></thead><tbody><p>42 rows × 3 columns</p><tr><th>1</th><td>20</td><td>1.5303</td><td>66</td></tr><tr><th>2</th><td>21</td><td>1.76271</td><td>118</td></tr><tr><th>3</th><td>22</td><td>2.05333</td><td>150</td></tr><tr><th>4</th><td>23</td><td>2.27536</td><td>138</td></tr><tr><th>5</th><td>24</td><td>2.25882</td><td>170</td></tr><tr><th>6</th><td>25</td><td>2.6092</td><td>174</td></tr><tr><th>7</th><td>26</td><td>2.67526</td><td>194</td></tr><tr><th>8</th><td>27</td><td>3.24138</td><td>203</td></tr><tr><th>9</th><td>28</td><td>3.33158</td><td>190</td></tr><tr><th>10</th><td>29</td><td>3.57754</td><td>187</td></tr><tr><th>11</th><td>30</td><td>3.96117</td><td>206</td></tr><tr><th>12</th><td>31</td><td>4.15169</td><td>178</td></tr><tr><th>13</th><td>32</td><td>4.62963</td><td>162</td></tr><tr><th>14</th><td>33</td><td>4.54491</td><td>167</td></tr><tr><th>15</th><td>34</td><td>4.39375</td><td>160</td></tr><tr><th>16</th><td>35</td><td>5.34078</td><td>179</td></tr><tr><th>17</th><td>36</td><td>5.42012</td><td>169</td></tr><tr><th>18</th><td>37</td><td>5.46341</td><td>164</td></tr><tr><th>19</th><td>38</td><td>5.72441</td><td>127</td></tr><tr><th>20</th><td>39</td><td>7.13115</td><td>122</td></tr><tr><th>21</th><td>40</td><td>7.2</td><td>125</td></tr><tr><th>22</th><td>41</td><td>6.21053</td><td>114</td></tr><tr><th>23</th><td>42</td><td>7.54918</td><td>122</td></tr><tr><th>24</th><td>43</td><td>6.9375</td><td>112</td></tr><tr><th>25</th><td>44</td><td>7.38043</td><td>92</td></tr><tr><th>26</th><td>45</td><td>7.46809</td><td>94</td></tr><tr><th>27</th><td>46</td><td>7.5</td><td>74</td></tr><tr><th>28</th><td>47</td><td>8.47727</td><td>88</td></tr><tr><th>29</th><td>48</td><td>9.45833</td><td>72</td></tr><tr><th>30</th><td>49</td><td>7.8806</td><td>67</td></tr><tr><th>&vellip;</th><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td></tr></tbody></table>"
      ],
      "text/latex": [
       "\\begin{tabular}{r|ccc}\n",
       "\t& age & mean\\_tenure & count\\\\\n",
       "\t\\hline\n",
       "\t& Int64⍰ & Float64 & Int64\\\\\n",
       "\t\\hline\n",
       "\t1 & 20 & 1.5303 & 66 \\\\\n",
       "\t2 & 21 & 1.76271 & 118 \\\\\n",
       "\t3 & 22 & 2.05333 & 150 \\\\\n",
       "\t4 & 23 & 2.27536 & 138 \\\\\n",
       "\t5 & 24 & 2.25882 & 170 \\\\\n",
       "\t6 & 25 & 2.6092 & 174 \\\\\n",
       "\t7 & 26 & 2.67526 & 194 \\\\\n",
       "\t8 & 27 & 3.24138 & 203 \\\\\n",
       "\t9 & 28 & 3.33158 & 190 \\\\\n",
       "\t10 & 29 & 3.57754 & 187 \\\\\n",
       "\t11 & 30 & 3.96117 & 206 \\\\\n",
       "\t12 & 31 & 4.15169 & 178 \\\\\n",
       "\t13 & 32 & 4.62963 & 162 \\\\\n",
       "\t14 & 33 & 4.54491 & 167 \\\\\n",
       "\t15 & 34 & 4.39375 & 160 \\\\\n",
       "\t16 & 35 & 5.34078 & 179 \\\\\n",
       "\t17 & 36 & 5.42012 & 169 \\\\\n",
       "\t18 & 37 & 5.46341 & 164 \\\\\n",
       "\t19 & 38 & 5.72441 & 127 \\\\\n",
       "\t20 & 39 & 7.13115 & 122 \\\\\n",
       "\t21 & 40 & 7.2 & 125 \\\\\n",
       "\t22 & 41 & 6.21053 & 114 \\\\\n",
       "\t23 & 42 & 7.54918 & 122 \\\\\n",
       "\t24 & 43 & 6.9375 & 112 \\\\\n",
       "\t25 & 44 & 7.38043 & 92 \\\\\n",
       "\t26 & 45 & 7.46809 & 94 \\\\\n",
       "\t27 & 46 & 7.5 & 74 \\\\\n",
       "\t28 & 47 & 8.47727 & 88 \\\\\n",
       "\t29 & 48 & 9.45833 & 72 \\\\\n",
       "\t30 & 49 & 7.8806 & 67 \\\\\n",
       "\t$\\dots$ & $\\dots$ & $\\dots$ & $\\dots$ \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "42×3 DataFrame\n",
       "│ Row │ age    │ mean_tenure │ count │\n",
       "│     │ \u001b[90mInt64⍰\u001b[39m │ \u001b[90mFloat64\u001b[39m     │ \u001b[90mInt64\u001b[39m │\n",
       "├─────┼────────┼─────────────┼───────┤\n",
       "│ 1   │ 20     │ 1.5303      │ 66    │\n",
       "│ 2   │ 21     │ 1.76271     │ 118   │\n",
       "│ 3   │ 22     │ 2.05333     │ 150   │\n",
       "│ 4   │ 23     │ 2.27536     │ 138   │\n",
       "│ 5   │ 24     │ 2.25882     │ 170   │\n",
       "│ 6   │ 25     │ 2.6092      │ 174   │\n",
       "│ 7   │ 26     │ 2.67526     │ 194   │\n",
       "│ 8   │ 27     │ 3.24138     │ 203   │\n",
       "│ 9   │ 28     │ 3.33158     │ 190   │\n",
       "│ 10  │ 29     │ 3.57754     │ 187   │\n",
       "⋮\n",
       "│ 32  │ 51     │ 9.24074     │ 54    │\n",
       "│ 33  │ 52     │ 10.5714     │ 77    │\n",
       "│ 34  │ 53     │ 8.78261     │ 46    │\n",
       "│ 35  │ 54     │ 11.5098     │ 51    │\n",
       "│ 36  │ 55     │ 12.4262     │ 61    │\n",
       "│ 37  │ 56     │ 12.3768     │ 69    │\n",
       "│ 38  │ 57     │ 10.8507     │ 67    │\n",
       "│ 39  │ 58     │ 14.5849     │ 53    │\n",
       "│ 40  │ 59     │ 13.814      │ 43    │\n",
       "│ 41  │ 60     │ 12.0702     │ 57    │\n",
       "│ 42  │ 61     │ 15.0476     │ 42    │"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tenure_by_age = by(benefits, :age, mean_tenure=:tenure=>mean, count=:age=>length, sort=true)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "Figure(PyObject <Figure size 640x480 with 1 Axes>)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "1-element Array{PyCall.PyObject,1}:\n",
       " PyObject <matplotlib.lines.Line2D object at 0x0000000001110908>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plot(tenure_by_age.age, tenure_by_age.mean_tenure)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "Figure(PyObject <Figure size 640x480 with 1 Axes>)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "1-element Array{PyCall.PyObject,1}:\n",
       " PyObject <matplotlib.lines.Line2D object at 0x00000000012E1400>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plot(tenure_by_age.age, tenure_by_age.count)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "sim_tenure (generic function with 1 method)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mutable struct Agent\n",
    "    age::Int\n",
    "    tenure::Int\n",
    "end\n",
    "\n",
    "function sim_tenure(p_lose::Float64, p_get::Float64, init_job::Float64, batch_size::Int)\n",
    "    agents = Agent[]\n",
    "    for i in 1:40\n",
    "        for a in agents\n",
    "            a.age += 1\n",
    "            if a.tenure == 0\n",
    "                a.tenure = rand() < p_lose ? 1 : 0\n",
    "            else\n",
    "                a.tenure = rand() < p_get ? 0 : a.tenure + 1\n",
    "            end\n",
    "        end\n",
    "        append!(agents, [Agent(20, rand() > init_job) for i in 1:batch_size])\n",
    "    end\n",
    "    DataFrame(agents)\n",
    "end"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"data-frame\"><thead><tr><th></th><th>age</th><th>tenure</th></tr><tr><th></th><th>Int64</th><th>Int64</th></tr></thead><tbody><p>4,000,000 rows × 2 columns</p><tr><th>1</th><td>59</td><td>18</td></tr><tr><th>2</th><td>59</td><td>9</td></tr><tr><th>3</th><td>59</td><td>0</td></tr><tr><th>4</th><td>59</td><td>0</td></tr><tr><th>5</th><td>59</td><td>9</td></tr><tr><th>6</th><td>59</td><td>6</td></tr><tr><th>7</th><td>59</td><td>7</td></tr><tr><th>8</th><td>59</td><td>0</td></tr><tr><th>9</th><td>59</td><td>7</td></tr><tr><th>10</th><td>59</td><td>13</td></tr><tr><th>11</th><td>59</td><td>0</td></tr><tr><th>12</th><td>59</td><td>0</td></tr><tr><th>13</th><td>59</td><td>10</td></tr><tr><th>14</th><td>59</td><td>0</td></tr><tr><th>15</th><td>59</td><td>0</td></tr><tr><th>16</th><td>59</td><td>0</td></tr><tr><th>17</th><td>59</td><td>0</td></tr><tr><th>18</th><td>59</td><td>2</td></tr><tr><th>19</th><td>59</td><td>23</td></tr><tr><th>20</th><td>59</td><td>0</td></tr><tr><th>21</th><td>59</td><td>19</td></tr><tr><th>22</th><td>59</td><td>0</td></tr><tr><th>23</th><td>59</td><td>7</td></tr><tr><th>24</th><td>59</td><td>0</td></tr><tr><th>25</th><td>59</td><td>0</td></tr><tr><th>26</th><td>59</td><td>9</td></tr><tr><th>27</th><td>59</td><td>0</td></tr><tr><th>28</th><td>59</td><td>0</td></tr><tr><th>29</th><td>59</td><td>7</td></tr><tr><th>30</th><td>59</td><td>7</td></tr><tr><th>&vellip;</th><td>&vellip;</td><td>&vellip;</td></tr></tbody></table>"
      ],
      "text/latex": [
       "\\begin{tabular}{r|cc}\n",
       "\t& age & tenure\\\\\n",
       "\t\\hline\n",
       "\t& Int64 & Int64\\\\\n",
       "\t\\hline\n",
       "\t1 & 59 & 18 \\\\\n",
       "\t2 & 59 & 9 \\\\\n",
       "\t3 & 59 & 0 \\\\\n",
       "\t4 & 59 & 0 \\\\\n",
       "\t5 & 59 & 9 \\\\\n",
       "\t6 & 59 & 6 \\\\\n",
       "\t7 & 59 & 7 \\\\\n",
       "\t8 & 59 & 0 \\\\\n",
       "\t9 & 59 & 7 \\\\\n",
       "\t10 & 59 & 13 \\\\\n",
       "\t11 & 59 & 0 \\\\\n",
       "\t12 & 59 & 0 \\\\\n",
       "\t13 & 59 & 10 \\\\\n",
       "\t14 & 59 & 0 \\\\\n",
       "\t15 & 59 & 0 \\\\\n",
       "\t16 & 59 & 0 \\\\\n",
       "\t17 & 59 & 0 \\\\\n",
       "\t18 & 59 & 2 \\\\\n",
       "\t19 & 59 & 23 \\\\\n",
       "\t20 & 59 & 0 \\\\\n",
       "\t21 & 59 & 19 \\\\\n",
       "\t22 & 59 & 0 \\\\\n",
       "\t23 & 59 & 7 \\\\\n",
       "\t24 & 59 & 0 \\\\\n",
       "\t25 & 59 & 0 \\\\\n",
       "\t26 & 59 & 9 \\\\\n",
       "\t27 & 59 & 0 \\\\\n",
       "\t28 & 59 & 0 \\\\\n",
       "\t29 & 59 & 7 \\\\\n",
       "\t30 & 59 & 7 \\\\\n",
       "\t$\\dots$ & $\\dots$ & $\\dots$ \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "4000000×2 DataFrame\n",
       "│ Row     │ age   │ tenure │\n",
       "│         │ \u001b[90mInt64\u001b[39m │ \u001b[90mInt64\u001b[39m  │\n",
       "├─────────┼───────┼────────┤\n",
       "│ 1       │ 59    │ 18     │\n",
       "│ 2       │ 59    │ 9      │\n",
       "│ 3       │ 59    │ 0      │\n",
       "│ 4       │ 59    │ 0      │\n",
       "│ 5       │ 59    │ 9      │\n",
       "│ 6       │ 59    │ 6      │\n",
       "│ 7       │ 59    │ 7      │\n",
       "│ 8       │ 59    │ 0      │\n",
       "│ 9       │ 59    │ 7      │\n",
       "│ 10      │ 59    │ 13     │\n",
       "⋮\n",
       "│ 3999990 │ 20    │ 1      │\n",
       "│ 3999991 │ 20    │ 0      │\n",
       "│ 3999992 │ 20    │ 1      │\n",
       "│ 3999993 │ 20    │ 0      │\n",
       "│ 3999994 │ 20    │ 1      │\n",
       "│ 3999995 │ 20    │ 1      │\n",
       "│ 3999996 │ 20    │ 0      │\n",
       "│ 3999997 │ 20    │ 0      │\n",
       "│ 3999998 │ 20    │ 0      │\n",
       "│ 3999999 │ 20    │ 0      │\n",
       "│ 4000000 │ 20    │ 1      │"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_sim_tenure = sim_tenure(0.1, 0.1, 0.5, 100000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"data-frame\"><thead><tr><th></th><th>age</th><th>job</th></tr><tr><th></th><th>Int64</th><th>Float64</th></tr></thead><tbody><p>40 rows × 2 columns</p><tr><th>1</th><td>59</td><td>0.50028</td></tr><tr><th>2</th><td>58</td><td>0.49714</td></tr><tr><th>3</th><td>57</td><td>0.50133</td></tr><tr><th>4</th><td>56</td><td>0.50016</td></tr><tr><th>5</th><td>55</td><td>0.49656</td></tr><tr><th>6</th><td>54</td><td>0.498</td></tr><tr><th>7</th><td>53</td><td>0.49785</td></tr><tr><th>8</th><td>52</td><td>0.50017</td></tr><tr><th>9</th><td>51</td><td>0.5004</td></tr><tr><th>10</th><td>50</td><td>0.50004</td></tr><tr><th>11</th><td>49</td><td>0.49894</td></tr><tr><th>12</th><td>48</td><td>0.50035</td></tr><tr><th>13</th><td>47</td><td>0.50007</td></tr><tr><th>14</th><td>46</td><td>0.49831</td></tr><tr><th>15</th><td>45</td><td>0.50173</td></tr><tr><th>16</th><td>44</td><td>0.50021</td></tr><tr><th>17</th><td>43</td><td>0.5004</td></tr><tr><th>18</th><td>42</td><td>0.49947</td></tr><tr><th>19</th><td>41</td><td>0.50098</td></tr><tr><th>20</th><td>40</td><td>0.50137</td></tr><tr><th>21</th><td>39</td><td>0.49881</td></tr><tr><th>22</th><td>38</td><td>0.50188</td></tr><tr><th>23</th><td>37</td><td>0.4986</td></tr><tr><th>24</th><td>36</td><td>0.50191</td></tr><tr><th>25</th><td>35</td><td>0.49826</td></tr><tr><th>26</th><td>34</td><td>0.50067</td></tr><tr><th>27</th><td>33</td><td>0.49994</td></tr><tr><th>28</th><td>32</td><td>0.49993</td></tr><tr><th>29</th><td>31</td><td>0.50137</td></tr><tr><th>30</th><td>30</td><td>0.50014</td></tr><tr><th>&vellip;</th><td>&vellip;</td><td>&vellip;</td></tr></tbody></table>"
      ],
      "text/latex": [
       "\\begin{tabular}{r|cc}\n",
       "\t& age & job\\\\\n",
       "\t\\hline\n",
       "\t& Int64 & Float64\\\\\n",
       "\t\\hline\n",
       "\t1 & 59 & 0.50028 \\\\\n",
       "\t2 & 58 & 0.49714 \\\\\n",
       "\t3 & 57 & 0.50133 \\\\\n",
       "\t4 & 56 & 0.50016 \\\\\n",
       "\t5 & 55 & 0.49656 \\\\\n",
       "\t6 & 54 & 0.498 \\\\\n",
       "\t7 & 53 & 0.49785 \\\\\n",
       "\t8 & 52 & 0.50017 \\\\\n",
       "\t9 & 51 & 0.5004 \\\\\n",
       "\t10 & 50 & 0.50004 \\\\\n",
       "\t11 & 49 & 0.49894 \\\\\n",
       "\t12 & 48 & 0.50035 \\\\\n",
       "\t13 & 47 & 0.50007 \\\\\n",
       "\t14 & 46 & 0.49831 \\\\\n",
       "\t15 & 45 & 0.50173 \\\\\n",
       "\t16 & 44 & 0.50021 \\\\\n",
       "\t17 & 43 & 0.5004 \\\\\n",
       "\t18 & 42 & 0.49947 \\\\\n",
       "\t19 & 41 & 0.50098 \\\\\n",
       "\t20 & 40 & 0.50137 \\\\\n",
       "\t21 & 39 & 0.49881 \\\\\n",
       "\t22 & 38 & 0.50188 \\\\\n",
       "\t23 & 37 & 0.4986 \\\\\n",
       "\t24 & 36 & 0.50191 \\\\\n",
       "\t25 & 35 & 0.49826 \\\\\n",
       "\t26 & 34 & 0.50067 \\\\\n",
       "\t27 & 33 & 0.49994 \\\\\n",
       "\t28 & 32 & 0.49993 \\\\\n",
       "\t29 & 31 & 0.50137 \\\\\n",
       "\t30 & 30 & 0.50014 \\\\\n",
       "\t$\\dots$ & $\\dots$ & $\\dots$ \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "40×2 DataFrame\n",
       "│ Row │ age   │ job     │\n",
       "│     │ \u001b[90mInt64\u001b[39m │ \u001b[90mFloat64\u001b[39m │\n",
       "├─────┼───────┼─────────┤\n",
       "│ 1   │ 59    │ 0.50028 │\n",
       "│ 2   │ 58    │ 0.49714 │\n",
       "│ 3   │ 57    │ 0.50133 │\n",
       "│ 4   │ 56    │ 0.50016 │\n",
       "│ 5   │ 55    │ 0.49656 │\n",
       "│ 6   │ 54    │ 0.498   │\n",
       "│ 7   │ 53    │ 0.49785 │\n",
       "│ 8   │ 52    │ 0.50017 │\n",
       "│ 9   │ 51    │ 0.5004  │\n",
       "│ 10  │ 50    │ 0.50004 │\n",
       "⋮\n",
       "│ 30  │ 30    │ 0.50014 │\n",
       "│ 31  │ 29    │ 0.49972 │\n",
       "│ 32  │ 28    │ 0.50425 │\n",
       "│ 33  │ 27    │ 0.49912 │\n",
       "│ 34  │ 26    │ 0.49908 │\n",
       "│ 35  │ 25    │ 0.50034 │\n",
       "│ 36  │ 24    │ 0.50096 │\n",
       "│ 37  │ 23    │ 0.50125 │\n",
       "│ 38  │ 22    │ 0.50285 │\n",
       "│ 39  │ 21    │ 0.49949 │\n",
       "│ 40  │ 20    │ 0.49902 │"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "job_by_age = by(df_sim_tenure, :age, job=:tenure=>x->mean(x .== 0))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "Figure(PyObject <Figure size 640x480 with 1 Axes>)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "1-element Array{PyCall.PyObject,1}:\n",
       " PyObject <matplotlib.lines.Line2D object at 0x0000000001143668>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plot(job_by_age.age, job_by_age.job)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"data-frame\"><thead><tr><th></th><th>age</th><th>mean_tenure</th><th>count</th></tr><tr><th></th><th>Int64</th><th>Float64</th><th>Int64</th></tr></thead><tbody><p>40 rows × 3 columns</p><tr><th>1</th><td>20</td><td>1.0</td><td>50098</td></tr><tr><th>2</th><td>21</td><td>1.89912</td><td>50051</td></tr><tr><th>3</th><td>22</td><td>2.71131</td><td>49715</td></tr><tr><th>4</th><td>23</td><td>3.43783</td><td>49875</td></tr><tr><th>5</th><td>24</td><td>4.0951</td><td>49904</td></tr><tr><th>6</th><td>25</td><td>4.67088</td><td>49966</td></tr><tr><th>7</th><td>26</td><td>5.23267</td><td>50092</td></tr><tr><th>8</th><td>27</td><td>5.68711</td><td>50088</td></tr><tr><th>9</th><td>28</td><td>6.14374</td><td>49575</td></tr><tr><th>10</th><td>29</td><td>6.52585</td><td>50028</td></tr><tr><th>11</th><td>30</td><td>6.85376</td><td>49986</td></tr><tr><th>12</th><td>31</td><td>7.17596</td><td>49863</td></tr><tr><th>13</th><td>32</td><td>7.51289</td><td>50007</td></tr><tr><th>14</th><td>33</td><td>7.71085</td><td>50006</td></tr><tr><th>15</th><td>34</td><td>7.97104</td><td>49933</td></tr><tr><th>16</th><td>35</td><td>8.15939</td><td>50174</td></tr><tr><th>17</th><td>36</td><td>8.39583</td><td>49809</td></tr><tr><th>18</th><td>37</td><td>8.46947</td><td>50140</td></tr><tr><th>19</th><td>38</td><td>8.63172</td><td>49812</td></tr><tr><th>20</th><td>39</td><td>8.84383</td><td>50119</td></tr><tr><th>21</th><td>40</td><td>8.87219</td><td>49863</td></tr><tr><th>22</th><td>41</td><td>8.9686</td><td>49902</td></tr><tr><th>23</th><td>42</td><td>9.14305</td><td>50053</td></tr><tr><th>24</th><td>43</td><td>9.18189</td><td>49960</td></tr><tr><th>25</th><td>44</td><td>9.2975</td><td>49979</td></tr><tr><th>26</th><td>45</td><td>9.37271</td><td>49827</td></tr><tr><th>27</th><td>46</td><td>9.40979</td><td>50169</td></tr><tr><th>28</th><td>47</td><td>9.43324</td><td>49993</td></tr><tr><th>29</th><td>48</td><td>9.53177</td><td>49965</td></tr><tr><th>30</th><td>49</td><td>9.55921</td><td>50106</td></tr><tr><th>&vellip;</th><td>&vellip;</td><td>&vellip;</td><td>&vellip;</td></tr></tbody></table>"
      ],
      "text/latex": [
       "\\begin{tabular}{r|ccc}\n",
       "\t& age & mean\\_tenure & count\\\\\n",
       "\t\\hline\n",
       "\t& Int64 & Float64 & Int64\\\\\n",
       "\t\\hline\n",
       "\t1 & 20 & 1.0 & 50098 \\\\\n",
       "\t2 & 21 & 1.89912 & 50051 \\\\\n",
       "\t3 & 22 & 2.71131 & 49715 \\\\\n",
       "\t4 & 23 & 3.43783 & 49875 \\\\\n",
       "\t5 & 24 & 4.0951 & 49904 \\\\\n",
       "\t6 & 25 & 4.67088 & 49966 \\\\\n",
       "\t7 & 26 & 5.23267 & 50092 \\\\\n",
       "\t8 & 27 & 5.68711 & 50088 \\\\\n",
       "\t9 & 28 & 6.14374 & 49575 \\\\\n",
       "\t10 & 29 & 6.52585 & 50028 \\\\\n",
       "\t11 & 30 & 6.85376 & 49986 \\\\\n",
       "\t12 & 31 & 7.17596 & 49863 \\\\\n",
       "\t13 & 32 & 7.51289 & 50007 \\\\\n",
       "\t14 & 33 & 7.71085 & 50006 \\\\\n",
       "\t15 & 34 & 7.97104 & 49933 \\\\\n",
       "\t16 & 35 & 8.15939 & 50174 \\\\\n",
       "\t17 & 36 & 8.39583 & 49809 \\\\\n",
       "\t18 & 37 & 8.46947 & 50140 \\\\\n",
       "\t19 & 38 & 8.63172 & 49812 \\\\\n",
       "\t20 & 39 & 8.84383 & 50119 \\\\\n",
       "\t21 & 40 & 8.87219 & 49863 \\\\\n",
       "\t22 & 41 & 8.9686 & 49902 \\\\\n",
       "\t23 & 42 & 9.14305 & 50053 \\\\\n",
       "\t24 & 43 & 9.18189 & 49960 \\\\\n",
       "\t25 & 44 & 9.2975 & 49979 \\\\\n",
       "\t26 & 45 & 9.37271 & 49827 \\\\\n",
       "\t27 & 46 & 9.40979 & 50169 \\\\\n",
       "\t28 & 47 & 9.43324 & 49993 \\\\\n",
       "\t29 & 48 & 9.53177 & 49965 \\\\\n",
       "\t30 & 49 & 9.55921 & 50106 \\\\\n",
       "\t$\\dots$ & $\\dots$ & $\\dots$ & $\\dots$ \\\\\n",
       "\\end{tabular}\n"
      ],
      "text/plain": [
       "40×3 DataFrame\n",
       "│ Row │ age   │ mean_tenure │ count │\n",
       "│     │ \u001b[90mInt64\u001b[39m │ \u001b[90mFloat64\u001b[39m     │ \u001b[90mInt64\u001b[39m │\n",
       "├─────┼───────┼─────────────┼───────┤\n",
       "│ 1   │ 20    │ 1.0         │ 50098 │\n",
       "│ 2   │ 21    │ 1.89912     │ 50051 │\n",
       "│ 3   │ 22    │ 2.71131     │ 49715 │\n",
       "│ 4   │ 23    │ 3.43783     │ 49875 │\n",
       "│ 5   │ 24    │ 4.0951      │ 49904 │\n",
       "│ 6   │ 25    │ 4.67088     │ 49966 │\n",
       "│ 7   │ 26    │ 5.23267     │ 50092 │\n",
       "│ 8   │ 27    │ 5.68711     │ 50088 │\n",
       "│ 9   │ 28    │ 6.14374     │ 49575 │\n",
       "│ 10  │ 29    │ 6.52585     │ 50028 │\n",
       "⋮\n",
       "│ 30  │ 49    │ 9.55921     │ 50106 │\n",
       "│ 31  │ 50    │ 9.59741     │ 49996 │\n",
       "│ 32  │ 51    │ 9.61801     │ 49960 │\n",
       "│ 33  │ 52    │ 9.66118     │ 49983 │\n",
       "│ 34  │ 53    │ 9.77063     │ 50215 │\n",
       "│ 35  │ 54    │ 9.77006     │ 50200 │\n",
       "│ 36  │ 55    │ 9.79598     │ 50344 │\n",
       "│ 37  │ 56    │ 9.80608     │ 49984 │\n",
       "│ 38  │ 57    │ 9.82235     │ 49867 │\n",
       "│ 39  │ 58    │ 9.7935      │ 50286 │\n",
       "│ 40  │ 59    │ 9.86428     │ 49972 │"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sim_tenure_by_age = by(filter(row -> row.tenure > 0, df_sim_tenure), :age,\n",
    "                       mean_tenure=:tenure=>mean, count=:age=>length, sort=true)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "Figure(PyObject <Figure size 640x480 with 1 Axes>)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "1-element Array{PyCall.PyObject,1}:\n",
       " PyObject <matplotlib.lines.Line2D object at 0x00000000011A7D30>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plot(sim_tenure_by_age.age, sim_tenure_by_age.mean_tenure)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "Figure(PyObject <Figure size 640x480 with 1 Axes>)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "1-element Array{PyCall.PyObject,1}:\n",
       " PyObject <matplotlib.lines.Line2D object at 0x000000000120F588>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plot(sim_tenure_by_age.age, sim_tenure_by_age.count)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Julia 1.1.0",
   "language": "julia",
   "name": "julia-1.1"
  },
  "language_info": {
   "file_extension": ".jl",
   "mimetype": "application/julia",
   "name": "julia",
   "version": "1.1.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}