rent_projections_v2_CLEAN.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Project rent burden for 2020-2021 school year\n",
    "\n",
    "This notebook estimates the change in number of on-campus grad students who will be rent burdened during the upcoming 2020-2021 school year following UCSD's proposed rent increases. This analysis makes the simplifying assumption that all students currently living in an on-campus unit will still be living in that unit next year. \n",
    "\n",
    "This analyis was performed well in advance of the COVID-19 shutdowns on campus and wider society, and as I document this notebook now on March 14th 2020, housing situation changes related to COVID-19 cannot be predicted or anticipated. Future versions of this analysis may be conducted to reflect any major changes in UCSD housing policy.\n",
    "\n",
    "**Note**: This notebook originally used the uncensored data. There may be slight differences when this analysis is reproduced with the public data. See the notebook in `data/` for more information on what was censored from the public release."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import plotly.graph_objs as go\n",
    "from scipy.stats import mannwhitneyu"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Check that pandas version is $\\geq$ 1.0."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'1.0.1'"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.__version__"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Check that plotly version is 4.5.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'4.5.0'"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import plotly\n",
    "plotly.__version__"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Load survey data\n",
    "\n",
    "This file is the raw survey data from Google Forms with the column names altered to be shorter."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>start_date</th>\n",
       "      <th>end_date</th>\n",
       "      <th>resp_type</th>\n",
       "      <th>ip</th>\n",
       "      <th>progress</th>\n",
       "      <th>duration</th>\n",
       "      <th>finished</th>\n",
       "      <th>rec_date</th>\n",
       "      <th>resp_id</th>\n",
       "      <th>...</th>\n",
       "      <th>n_dependents</th>\n",
       "      <th>has_healthcare</th>\n",
       "      <th>dependent_has_healthcare_CAT</th>\n",
       "      <th>depdendent_has_healthcare_TXT</th>\n",
       "      <th>expenses_CAT</th>\n",
       "      <th>expenses_TXT</th>\n",
       "      <th>expenses_cost_month</th>\n",
       "      <th>comments</th>\n",
       "      <th>parent_topics</th>\n",
       "      <th>topics</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>2020-02-05 16:42:04</td>\n",
       "      <td>2020-02-05 16:45:55</td>\n",
       "      <td>Survey Preview</td>\n",
       "      <td>NaN</td>\n",
       "      <td>100</td>\n",
       "      <td>230</td>\n",
       "      <td>True</td>\n",
       "      <td>2020-02-05 16:45:55.453000</td>\n",
       "      <td>R_2ir9mTnzPwy5Dnb</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>You are a goddess</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Unknown</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>2020-02-05 18:27:44</td>\n",
       "      <td>2020-02-05 19:10:56</td>\n",
       "      <td>Survey Preview</td>\n",
       "      <td>NaN</td>\n",
       "      <td>100</td>\n",
       "      <td>2591</td>\n",
       "      <td>True</td>\n",
       "      <td>2020-02-05 19:10:56.668000</td>\n",
       "      <td>R_10TRlQ4wnfjELpB</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Unknown</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>2020-02-05 19:09:12</td>\n",
       "      <td>2020-02-05 19:12:38</td>\n",
       "      <td>Survey Preview</td>\n",
       "      <td>NaN</td>\n",
       "      <td>100</td>\n",
       "      <td>205</td>\n",
       "      <td>True</td>\n",
       "      <td>2020-02-05 19:12:38.293000</td>\n",
       "      <td>R_V3IqukjLVB20BLX</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Unknown</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>2020-02-05 19:09:36</td>\n",
       "      <td>2020-02-05 19:12:42</td>\n",
       "      <td>Survey Preview</td>\n",
       "      <td>NaN</td>\n",
       "      <td>100</td>\n",
       "      <td>186</td>\n",
       "      <td>True</td>\n",
       "      <td>2020-02-05 19:12:43.306000</td>\n",
       "      <td>R_Zt6ip9HGmdFeHrX</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Unknown</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>2020-02-05 19:10:32</td>\n",
       "      <td>2020-02-05 19:12:51</td>\n",
       "      <td>Survey Preview</td>\n",
       "      <td>NaN</td>\n",
       "      <td>100</td>\n",
       "      <td>138</td>\n",
       "      <td>True</td>\n",
       "      <td>2020-02-05 19:12:51.718000</td>\n",
       "      <td>R_1jMJreLskBbuWpc</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 65 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0           start_date             end_date       resp_type   ip  \\\n",
       "0           1  2020-02-05 16:42:04  2020-02-05 16:45:55  Survey Preview  NaN   \n",
       "1           2  2020-02-05 18:27:44  2020-02-05 19:10:56  Survey Preview  NaN   \n",
       "2           3  2020-02-05 19:09:12  2020-02-05 19:12:38  Survey Preview  NaN   \n",
       "3           4  2020-02-05 19:09:36  2020-02-05 19:12:42  Survey Preview  NaN   \n",
       "4           5  2020-02-05 19:10:32  2020-02-05 19:12:51  Survey Preview  NaN   \n",
       "\n",
       "   progress  duration  finished                    rec_date  \\\n",
       "0       100       230      True  2020-02-05 16:45:55.453000   \n",
       "1       100      2591      True  2020-02-05 19:10:56.668000   \n",
       "2       100       205      True  2020-02-05 19:12:38.293000   \n",
       "3       100       186      True  2020-02-05 19:12:43.306000   \n",
       "4       100       138      True  2020-02-05 19:12:51.718000   \n",
       "\n",
       "             resp_id  ...  n_dependents  has_healthcare  \\\n",
       "0  R_2ir9mTnzPwy5Dnb  ...             1             NaN   \n",
       "1  R_10TRlQ4wnfjELpB  ...           NaN             NaN   \n",
       "2  R_V3IqukjLVB20BLX  ...             0             NaN   \n",
       "3  R_Zt6ip9HGmdFeHrX  ...             0             NaN   \n",
       "4  R_1jMJreLskBbuWpc  ...           NaN             NaN   \n",
       "\n",
       "   dependent_has_healthcare_CAT  depdendent_has_healthcare_TXT  expenses_CAT  \\\n",
       "0                           NaN                            NaN           NaN   \n",
       "1                           NaN                            NaN           NaN   \n",
       "2                           NaN                            NaN           NaN   \n",
       "3                           NaN                            NaN           NaN   \n",
       "4                           NaN                            NaN           NaN   \n",
       "\n",
       "   expenses_TXT expenses_cost_month           comments parent_topics   topics  \n",
       "0           NaN                 NaN  You are a goddess           NaN  Unknown  \n",
       "1           NaN                 NaN                NaN           NaN  Unknown  \n",
       "2           NaN                 NaN                NaN           NaN  Unknown  \n",
       "3           NaN                 NaN                NaN           NaN  Unknown  \n",
       "4           NaN                 NaN                NaN           NaN      NaN  \n",
       "\n",
       "[5 rows x 65 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv(\"2020_housingsurvey/clean_data/2020_15.50_short-cols.tsv\", sep = '\\t')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Clean dollar amounts in rent column\n",
    "\n",
    "This function cleans up the low-hanging fruit in reported dollar amounts. Values that could not be cleaned (converted to integers automatically) will eventually be excluded. I have written this function to print out these values so at the very least these reported experiences will be documented somewhere. \"Homeless\" is not an integer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def clean_dollars(x):\n",
    "    val = str(x)\n",
    "    try:\n",
    "        return int(val.split()[0].replace(\"$\",\"\").replace(\",\",\"\").replace(\"~\",\"\").split('.')[0])\n",
    "    except (TypeError, ValueError):\n",
    "        print(val)\n",
    "        return val"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Homeless\n",
      "I live with family because I can not afford rent\n",
      "850-870\n",
      "$pence 1074\n",
      "I have a mortgate, and two kids in college I pay both their rents too $5000\n",
      "Right now 600 but I will soon have to pay 1000\n",
      "personal: $1150; whole house: $3950\n"
     ]
    }
   ],
   "source": [
    "df[\"rent_month\"] = df[\"rent_month\"].apply(clean_dollars)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Clean dollar amounts in income column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1600/month\n",
      "Around $28000\n",
      "None\n",
      "not enough\n",
      "-\n",
      "Depends.  about $25,000 for 2 years fellowship, TA wages for other years\n",
      "32’000\n",
      "30k last fiscal year, according to W2\n",
      "Around 21,000\n",
      "￥1，000\n",
      "approximately $28,800\n",
      "nan\n",
      "first year: $4,000 in stipend/award, $7,000-14,000 employment salary range\n",
      "Program costs and variable employment\n",
      "32K\n",
      "~ $21,000\n",
      "?\n",
      "18000-24000\n",
      "nan\n",
      "$30K (pre-tax income)\n",
      "Only tuition remission \n",
      "Allegedly $33,000\n",
      "$26,000-27,000\n",
      "nan\n",
      "$ 13368\n",
      "Not guaranteed\n",
      "&lt;20,000\n",
      "about $21,000 before tax but after mandatory fees\n"
     ]
    }
   ],
   "source": [
    "df[\"grad_prog_comp\"] = df[\"grad_prog_comp\"].apply(clean_dollars)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Check both income and rent are integers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "def are_income_rent_ints(row):\n",
    "    return type(row[\"grad_prog_comp\"]) == int and type(row[\"rent_month\"]) == int"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(971, 65)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(936, 65)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(df.shape)\n",
    "\n",
    "df = df[df.apply(are_income_rent_ints, axis = 1)]\n",
    "\n",
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After cleaning and converting relevant dollar amounts to integers, 936 observations remain."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Drop rows where `grad_prog_comp` is 0\n",
    "\n",
    "Graduate program compensation is used as income. Therefore this column cannot be zero for rent burden calculations."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(936, 65)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(889, 65)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(df.shape)\n",
    "\n",
    "df = df[df[\"grad_prog_comp\"] > 0]\n",
    "\n",
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Calculate rent fraction of income"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"rent_burden\"] = (df[\"rent_month\"] * 12) / df[\"grad_prog_comp\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Remove rent burden values > 1.0\n",
    "\n",
    "These are assumed to be erroneous observations. Some respondents very likely included the rent for their entire unit, not just the room they pay for. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(889, 66)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(835, 66)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(df.shape)\n",
    "\n",
    "df = df[df[\"rent_burden\"] < 1.0]\n",
    "\n",
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Function: Violin w/rent burden line\n",
    "\n",
    "Plot the distribution of rent burdens from the survey."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_rent_burden(rent_burdens):\n",
    "    fig = go.Figure()\n",
    "    fig.add_trace(\n",
    "        go.Violin(\n",
    "            y = rent_burdens,\n",
    "            name = \"Rent Burden\",\n",
    "            box_visible = True,\n",
    "            meanline_visible = True\n",
    "        )\n",
    "    )\n",
    "    fig.update_layout(template='plotly_white')\n",
    "    fig.show(renderer=\"png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_rent_burden(df[\"rent_burden\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Visualize rent burden for on campus\n",
    "\n",
    "Visualize just rent burdens for on campus grad students."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "on_campus_categories = [\"On-campus (lease/rent)\", \"On-campus (sublet)\"]\n",
    "\n",
    "plot_rent_burden(\n",
    "    df[df[\"on_campus\"].isin(on_campus_categories)][\"rent_burden\"]\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Construct a dataframe of current and projected rents\n",
    "\n",
    "This uses Petia's data [here](https://docs.google.com/spreadsheets/d/1r1sWVa-RuEmTx7nQN08XOEJHT4Uqi7thAY3qwDDQ0Lg/edit#gid=0). Rent amounts are the amount *per room*, i.e., 2bed rates are divided by 2."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>unit_type</th>\n",
       "      <th>age</th>\n",
       "      <th>n_beds</th>\n",
       "      <th>rent_19-20</th>\n",
       "      <th>rent_20-21</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>coastst</td>\n",
       "      <td>old</td>\n",
       "      <td>21</td>\n",
       "      <td>794.0</td>\n",
       "      <td>818.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>coast1bd</td>\n",
       "      <td>old</td>\n",
       "      <td>54</td>\n",
       "      <td>1075.0</td>\n",
       "      <td>1108.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>coast2bd</td>\n",
       "      <td>old</td>\n",
       "      <td>62</td>\n",
       "      <td>751.0</td>\n",
       "      <td>773.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>cm1bd</td>\n",
       "      <td>old</td>\n",
       "      <td>9</td>\n",
       "      <td>1103.0</td>\n",
       "      <td>1136.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>cm2bd</td>\n",
       "      <td>old</td>\n",
       "      <td>364</td>\n",
       "      <td>608.5</td>\n",
       "      <td>627.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  unit_type  age  n_beds  rent_19-20  rent_20-21\n",
       "0   coastst  old      21       794.0       818.0\n",
       "1  coast1bd  old      54      1075.0      1108.0\n",
       "2  coast2bd  old      62       751.0       773.5\n",
       "3     cm1bd  old       9      1103.0      1136.0\n",
       "4     cm2bd  old     364       608.5       627.0"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "housing_df = pd.DataFrame([\n",
    "    (\"coastst\", \"old\", 21, 794, 818),\n",
    "    (\"coast1bd\", \"old\", 54, 1075, 1108),\n",
    "    (\"coast2bd\", \"old\", 62, 1502/2.0, 1547/2.0),\n",
    "    (\"cm1bd\", \"old\", 9, 1103, 1136),\n",
    "    (\"cm2bd\", \"old\", 364, 1217/2.0, 1254/2.0),\n",
    "    (\"cm3bd\", \"old\", 27, 1412/3.0, 1454/3.0),\n",
    "    (\"sm2bd\", \"old\", 380, 1409/2.0, 1451/2.0),\n",
    "    (\"oms2bd\", \"old\", 806, 1307/2.0, 1346/2.0),\n",
    "    (\"mnst\", \"old\", 853, 849, 900),\n",
    "    (\"mn1bd\", \"old\", 44, 1191, 1250),\n",
    "    (\"mn2bd\", \"old\", 332, 1666/2.0, 1750/2.0),\n",
    "    (\"mn3bd\", \"old\", 120, 2470/3.0, 2593/3.0),\n",
    "    (\"ne1bd\", \"new\", 199, np.nan, 1302),\n",
    "    (\"ne2b1b\", \"new\", 536, np.nan, 1851/2.0),\n",
    "    (\"ne3b2b\", \"new\", 324, np.nan, 2850/3.0),\n",
    "    (\"ne4b2b\", \"new\", 304, np.nan, 3600/4.0),\n",
    "    (\"nw2b2b\", \"new\", 76, np.nan, 2080/2.0),\n",
    "    (\"nw2b1b\", \"new\", 228, np.nan, 1930/2.0),\n",
    "    (\"nw4b2b\", \"new\", 528, np.nan, 3760/4.0),\n",
    "    (\"nw6b3b\", \"new\", 42, np.nan, 5490/6.0)\n",
    "])\n",
    "housing_df.columns = [\"unit_type\", \"age\", \"n_beds\", \"rent_19-20\", \"rent_20-21\"]\n",
    "housing_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Function: Convert row in survey to `unit_type` in `housing_df`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "building_map = {\n",
    "    \"Coast\": \"coast\",\n",
    "    \"Mesa\": \"cm\",\n",
    "    \"Mesa Nueva\": \"mn\",\n",
    "    \"One Miramar St.\": \"oms\"\n",
    "}\n",
    "\n",
    "size_map = {\n",
    "    \"Studio\": \"st\",\n",
    "    \"1 bedroom\": \"1bd\",\n",
    "    \"2 bedrooms\": \"2bd\",\n",
    "    \"3 bedrooms\": \"3bd\"\n",
    "}\n",
    "\n",
    "def get_unit_type(row):\n",
    "    try:\n",
    "        building = building_map[row[\"house_devel\"]]\n",
    "    except KeyError:\n",
    "        return np.nan\n",
    "    \n",
    "    try:\n",
    "        size = size_map[row[\"n_beds\"]]\n",
    "    except KeyError:\n",
    "        return np.nan\n",
    "    \n",
    "    return building+size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(358, 66)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/alexwenzel/miniconda3/lib/python3.7/site-packages/ipykernel_launcher.py:6: SettingWithCopyWarning:\n",
      "\n",
      "\n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(332, 67)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_on_campus = df[df[\"on_campus\"].isin([\"On-campus (lease/rent)\", \"On-campus (sublet)\"])]\n",
    "\n",
    "print(df_on_campus.shape)\n",
    "\n",
    "#df_on_campus[\"unit_type\"] = df_on_campus.apply(get_unit_type, axis = 1).dropna(subset = [\"unit_type\"])\n",
    "df_on_campus[\"unit_type\"] = df_on_campus.apply(get_unit_type, axis = 1)\n",
    "df_on_campus = df_on_campus.dropna(subset = [\"unit_type\"])\n",
    "\n",
    "df_on_campus.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Convert central mesa studio to 1 bd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_on_campus[\"unit_type\"] = df_on_campus[\"unit_type\"].replace(\"cmst\", \"cm1bd\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Map new rents\n",
    "\n",
    "Here the current rents from the surveys are projected to next year using the proposed increases."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "unit_type\n",
       "coastst      818.0\n",
       "coast1bd    1108.0\n",
       "coast2bd     773.5\n",
       "cm1bd       1136.0\n",
       "cm2bd        627.0\n",
       "Name: rent_20-21, dtype: float64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_rent_map = housing_df[\"rent_20-21\"]\n",
    "new_rent_map.index = housing_df[\"unit_type\"]\n",
    "new_rent_map.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_on_campus[\"new_rent_month\"] = df_on_campus.apply(lambda row: new_rent_map[row[\"unit_type\"]], axis = 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Calculate new rent burden"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_on_campus[\"new_rent_burden\"] = (df_on_campus[\"new_rent_month\"] * 12) / df_on_campus[\"grad_prog_comp\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_rent_burden(\n",
    "    df_on_campus[\"new_rent_burden\"]\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Final df before/after"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "final_oc = df_on_campus[\n",
    "    [\"grad_prog_comp\", \"rent_month\", \"rent_burden\", \"unit_type\", \"new_rent_month\", \"new_rent_burden\"]\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>grad_prog_comp</th>\n",
       "      <th>rent_month</th>\n",
       "      <th>rent_burden</th>\n",
       "      <th>unit_type</th>\n",
       "      <th>new_rent_month</th>\n",
       "      <th>new_rent_burden</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>20000</td>\n",
       "      <td>846</td>\n",
       "      <td>0.5076</td>\n",
       "      <td>mnst</td>\n",
       "      <td>900.000000</td>\n",
       "      <td>0.54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>33000</td>\n",
       "      <td>670</td>\n",
       "      <td>0.243636</td>\n",
       "      <td>cm2bd</td>\n",
       "      <td>627.000000</td>\n",
       "      <td>0.228</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>30000</td>\n",
       "      <td>849</td>\n",
       "      <td>0.3396</td>\n",
       "      <td>mnst</td>\n",
       "      <td>900.000000</td>\n",
       "      <td>0.36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>34000</td>\n",
       "      <td>653</td>\n",
       "      <td>0.230471</td>\n",
       "      <td>oms2bd</td>\n",
       "      <td>673.000000</td>\n",
       "      <td>0.237529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>34000</td>\n",
       "      <td>850</td>\n",
       "      <td>0.3</td>\n",
       "      <td>mn3bd</td>\n",
       "      <td>864.333333</td>\n",
       "      <td>0.305059</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  grad_prog_comp rent_month rent_burden unit_type  new_rent_month  \\\n",
       "1          20000        846      0.5076      mnst      900.000000   \n",
       "2          33000        670    0.243636     cm2bd      627.000000   \n",
       "4          30000        849      0.3396      mnst      900.000000   \n",
       "5          34000        653    0.230471    oms2bd      673.000000   \n",
       "8          34000        850         0.3     mn3bd      864.333333   \n",
       "\n",
       "  new_rent_burden  \n",
       "1            0.54  \n",
       "2           0.228  \n",
       "4            0.36  \n",
       "5        0.237529  \n",
       "8        0.305059  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final_oc.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Combined violin plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = go.Figure()\n",
    "\n",
    "greater = final_oc[final_oc[\"new_rent_month\"] > final_oc[\"rent_month\"]]\n",
    "\n",
    "curr_rent_burden = list(greater[\"rent_burden\"])\n",
    "proj_rent_burden = list(greater[\"new_rent_burden\"])\n",
    "\n",
    "for pair in [(\"curr_rent_burden\", curr_rent_burden), (\"proj_rent_burden\", proj_rent_burden)]:\n",
    "    fig.add_trace(\n",
    "        go.Violin(\n",
    "            x = [pair[0]]*len(pair[1]),\n",
    "            y = pair[1],\n",
    "            name = pair[0],\n",
    "            box_visible = True,\n",
    "            meanline_visible = True\n",
    "        )\n",
    "    )\n",
    "    \n",
    "fig.add_trace(\n",
    "    go.Scatter(\n",
    "        x = [\"curr_rent_burden\", \"proj_rent_burden\"],\n",
    "        y = [0.3, 0.3],\n",
    "        line = dict(\n",
    "            color = 'red'\n",
    "        )\n",
    "    )\n",
    ")\n",
    "\n",
    "fig.update_layout(template = 'plotly_white')\n",
    "fig.show(renderer=\"png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.6954545454545454"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "greater[greater[\"rent_burden\"]>0.3].shape[0]/greater.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8181818181818182"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "greater[greater[\"new_rent_burden\"]>0.3].shape[0]/greater.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(220, 6)"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "greater.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'cm1bd',\n",
       " 'cm2bd',\n",
       " 'coast1bd',\n",
       " 'coast2bd',\n",
       " 'coastst',\n",
       " 'mn1bd',\n",
       " 'mn2bd',\n",
       " 'mn3bd',\n",
       " 'mnst',\n",
       " 'oms2bd'}"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "set(greater[\"unit_type\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}