portfolio.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Are university towns housing prices less affected by recession?\n",
    "\n",
    "In this data analysis project, we will formulate a hypothesis around this question, gather and process required data and finally test our hypothesis.\n",
    "\n",
    "### Strategy\n",
    "\n",
    "We will use 3 datasources for this project. \n",
    "1. We will use GDP data from Bureau of Economic Analysis, US Department of Commerce, the GDP over time of the United States in current dollars (using the chained value in 2009 dollars), in quarterly intervals, in the file gdplev.xls. We will use this data to determine recession periods. For this project, we will only look at GDP data from the first quarter of 2000 onward.\n",
    "\n",
    "2. We will use housing price data from the Zillow research data site. There is housing data for the United States. In particular the datafile for all homes at a city level, 'City_Zhvi_AllHomes.csv', has median home sale prices at a fine grained level.\n",
    "\n",
    "3. We will use a list of university towns collected from Wikipedia to divide the housing price data in two sets as university towns and non-university towns. We have this data in the 'university_towns.txt' file.\n",
    "\n",
    "So, we will use the above mentioned data sources and manipulate and transform them to test our hypothesis.\n",
    "\n",
    "### Hypothesis\n",
    "#### University towns have their mean housing prices less affected by recessions.\n",
    "\n",
    "### Required Definitions\n",
    "* A _quarter_ is a specific three month period, Q1 is January through March, Q2 is April through June, Q3 is July through September, Q4 is October through December.\n",
    "* A _recession_ is defined as starting with two consecutive quarters of GDP decline, and ending with two consecutive quarters of GDP growth.\n",
    "* A _recession bottom_ is the quarter within a recession which had the lowest GDP.\n",
    "* A _university town_ is a city which has a high percentage of university students compared to the total population of the city.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "### Import necessary libraries\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from scipy.stats import ttest_ind\n",
    "\n",
    "from matplotlib import pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Use this dictionary to map state names to two letter acronyms\n",
    "states = {'OH': 'Ohio', 'KY': 'Kentucky', 'AS': 'American Samoa', 'NV': 'Nevada', 'WY': 'Wyoming', 'NA': 'National', 'AL': 'Alabama', 'MD': 'Maryland', 'AK': 'Alaska', 'UT': 'Utah', 'OR': 'Oregon', 'MT': 'Montana', 'IL': 'Illinois', 'TN': 'Tennessee', 'DC': 'District of Columbia', 'VT': 'Vermont', 'ID': 'Idaho', 'AR': 'Arkansas', 'ME': 'Maine', 'WA': 'Washington', 'HI': 'Hawaii', 'WI': 'Wisconsin', 'MI': 'Michigan', 'IN': 'Indiana', 'NJ': 'New Jersey', 'AZ': 'Arizona', 'GU': 'Guam', 'MS': 'Mississippi', 'PR': 'Puerto Rico', 'NC': 'North Carolina', 'TX': 'Texas', 'SD': 'South Dakota', 'MP': 'Northern Mariana Islands', 'IA': 'Iowa', 'MO': 'Missouri', 'CT': 'Connecticut', 'WV': 'West Virginia', 'SC': 'South Carolina', 'LA': 'Louisiana', 'KS': 'Kansas', 'NY': 'New York', 'NE': 'Nebraska', 'OK': 'Oklahoma', 'FL': 'Florida', 'CA': 'California', 'CO': 'Colorado', 'PA': 'Pennsylvania', 'DE': 'Delaware', 'NM': 'New Mexico', 'RI': 'Rhode Island', 'MN': 'Minnesota', 'VI': 'Virgin Islands', 'NH': 'New Hampshire', 'MA': 'Massachusetts', 'GA': 'Georgia', 'ND': 'North Dakota', 'VA': 'Virginia'}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def get_list_of_university_towns():\n",
    "    '''Returns a DataFrame of towns and the states they are in from the \n",
    "    university_towns.txt list. The format of the DataFrame should be:\n",
    "    DataFrame( [ [\"Michigan\", \"Ann Arbor\"], [\"Michigan\", \"Yipsilanti\"] ], \n",
    "    columns=[\"State\", \"RegionName\"]  )\n",
    "    \n",
    "    The following cleaning needs to be done:\n",
    "\n",
    "    1. For \"State\", removing characters from \"[\" to the end.\n",
    "    2. For \"RegionName\", when applicable, removing every character from \" (\" to the end.\n",
    "    3. Remove newline character '\\n'. '''\n",
    "    ut_list = []\n",
    "    \n",
    "    with open('university_towns.txt') as f:\n",
    "        for line in f:\n",
    "            if 'edit' in line:\n",
    "                current_city = line.split('[')[0].strip()\n",
    "            else:\n",
    "                ut_list.append((current_city, line.split('(')[0].strip()))\n",
    "                \n",
    "    ut_df = pd.DataFrame.from_records(ut_list)\n",
    "    ut_df.columns = ['State', 'RegionName']\n",
    "    return ut_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "scrolled": true
   },
   "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>State</th>\n",
       "      <th>RegionName</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Alabama</td>\n",
       "      <td>Auburn</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Alabama</td>\n",
       "      <td>Florence</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Alabama</td>\n",
       "      <td>Jacksonville</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Alabama</td>\n",
       "      <td>Livingston</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Alabama</td>\n",
       "      <td>Montevallo</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     State    RegionName\n",
       "0  Alabama        Auburn\n",
       "1  Alabama      Florence\n",
       "2  Alabama  Jacksonville\n",
       "3  Alabama    Livingston\n",
       "4  Alabama    Montevallo"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ut_df = get_list_of_university_towns()\n",
    "ut_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def get_gdp_df():\n",
    "    '''this function reads the gdp data and returns a dataframe with only the required columns'''\n",
    "    gdplev = pd.ExcelFile('gdplev.xls')\n",
    "    gdplev = gdplev.parse(\"Sheet1\", skiprows=219)\n",
    "    gdplev = gdplev[['1999q4', 9926.1]]\n",
    "    gdplev.columns = ['quarter','gdp']\n",
    "    return gdplev"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's look at the gdp data visually first."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "gdp = get_gdp_df()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABKoAAAKHCAYAAAChAqDBAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAgCFoVAEAAAAwhP8fgGc0U9n+qFgAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2272f25d080>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = gdp.plot.bar(figsize=(20, 10),  legend=True, x='quarter', y='gdp')\n",
    "plt.title('GDP in Quarters(From 2000 to 2016)', fontsize='22')\n",
    "plt.ylabel('GDP in Billions', fontsize='18')\n",
    "plt.xlabel('Quarters', fontsize='18')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " We can see there's small dip in the middle. Let's find out the recession period and look at it more closely."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def get_recession_start():\n",
    "    '''Returns the year and quarter of the recession start time as a \n",
    "    string value in a format such as 2005q3'''\n",
    "    gdplev = get_gdp_df()\n",
    "    for i in range(2, len(gdplev)):\n",
    "        if (gdplev.iloc[i-2][1] > gdplev.iloc[i-1][1]) and (gdplev.iloc[i-1][1] > gdplev.iloc[i][1]):\n",
    "            return gdplev.iloc[i-2][0]\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The above function finds the start of a recession based on the definition above. Let's run this and find the starting point for recession."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'2008q3'"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "get_recession_start()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So, within our window of analysis, a recession had started in the 3rd quarter of 2008 as is well known"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This function finds the end of a recession according to the definition."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def get_recession_end():\n",
    "    '''Returns the year and quarter of the recession end time as a \n",
    "    string value in a format such as 2005q3'''\n",
    "\n",
    "    gdplev = get_gdp_df()\n",
    "    start = get_recession_start()\n",
    "    start_index = gdplev[gdplev['quarter'] == start].index.tolist()[0]\n",
    "    gdplev=gdplev.iloc[start_index:]\n",
    "    for i in range(2, len(gdplev)):\n",
    "        if (gdplev.iloc[i-2][1] < gdplev.iloc[i-1][1]) and (gdplev.iloc[i-1][1] < gdplev.iloc[i][1]):\n",
    "            return gdplev.iloc[i][0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's run this and find out the end of the recession period"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'2009q4'"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "get_recession_end()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So, the recession ended in the last quarter of 2009. It lasted over a year!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now let's figure out the bottom of the recession"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def get_recession_bottom():\n",
    "    '''Returns the year and quarter of the recession bottom time as a \n",
    "    string value in a format such as 2005q3'''\n",
    "    gdp = get_gdp_df()\n",
    "    a = gdp.index[gdp['quarter'] == get_recession_start()][0]\n",
    "    b = gdp.index[gdp['quarter'] == get_recession_end()][0]\n",
    "    \n",
    "    \n",
    "    return gdp.iloc[gdp.iloc[a:b+1, 1 ].idxmin(), 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'2009q2'"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "get_recession_bottom()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, that we have figured out the recession period, let's visualize the period."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "recession_start_index = gdp.loc[gdp.quarter == get_recession_start(), :].index[0]\n",
    "recession_end_index = gdp.loc[gdp.quarter == get_recession_end(), :].index[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# we take two previous quarters before recession into account to capture the start of recession in the graph\n",
    "recession = gdp.iloc[recession_start_index-2: recession_end_index+1, :] "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "([<matplotlib.axis.XTick at 0x2272feb0588>,\n",
       "  <matplotlib.axis.XTick at 0x2272f8bbc88>,\n",
       "  <matplotlib.axis.XTick at 0x2272fb656a0>,\n",
       "  <matplotlib.axis.XTick at 0x2272fec3ef0>,\n",
       "  <matplotlib.axis.XTick at 0x2272feca588>,\n",
       "  <matplotlib.axis.XTick at 0x2272fecabe0>,\n",
       "  <matplotlib.axis.XTick at 0x2272fecf278>,\n",
       "  <matplotlib.axis.XTick at 0x2272fec3978>],\n",
       " <a list of 8 Text xticklabel objects>)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABKsAAAJuCAYAAAB/i4G1AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2272fb3f208>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "recession.set_index('quarter').plot.line(figsize=(20, 10), style='-', marker='o')\n",
    "plt.title('GDP in Quarters (Recession Period)', fontsize='24')\n",
    "plt.xlabel('Quarters', fontsize=\"20\")\n",
    "plt.ylabel('GDP in billions', fontsize='20')\n",
    "plt.xticks([x for x in range(len(recession.index))], recession.quarter)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So, the 2nd quarter of 2009 had the lowest GDP. We can see two consecutive decrease starting from 2008q3 and two consecutive growth starting from 2009q3. So, 2009q4 marks the end of this recession period."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So, we have determined the important data points related to recession. Let's look into the housing price data now."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>RegionID</th>\n",
       "      <th>RegionName</th>\n",
       "      <th>State</th>\n",
       "      <th>Metro</th>\n",
       "      <th>CountyName</th>\n",
       "      <th>SizeRank</th>\n",
       "      <th>1996-04</th>\n",
       "      <th>1996-05</th>\n",
       "      <th>1996-06</th>\n",
       "      <th>1996-07</th>\n",
       "      <th>...</th>\n",
       "      <th>2015-11</th>\n",
       "      <th>2015-12</th>\n",
       "      <th>2016-01</th>\n",
       "      <th>2016-02</th>\n",
       "      <th>2016-03</th>\n",
       "      <th>2016-04</th>\n",
       "      <th>2016-05</th>\n",
       "      <th>2016-06</th>\n",
       "      <th>2016-07</th>\n",
       "      <th>2016-08</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>6181</td>\n",
       "      <td>New York</td>\n",
       "      <td>NY</td>\n",
       "      <td>New York</td>\n",
       "      <td>Queens</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>573600</td>\n",
       "      <td>576200</td>\n",
       "      <td>578400</td>\n",
       "      <td>582200</td>\n",
       "      <td>588000</td>\n",
       "      <td>592200</td>\n",
       "      <td>592500</td>\n",
       "      <td>590200</td>\n",
       "      <td>588000</td>\n",
       "      <td>586400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>12447</td>\n",
       "      <td>Los Angeles</td>\n",
       "      <td>CA</td>\n",
       "      <td>Los Angeles-Long Beach-Anaheim</td>\n",
       "      <td>Los Angeles</td>\n",
       "      <td>2</td>\n",
       "      <td>155000.0</td>\n",
       "      <td>154600.0</td>\n",
       "      <td>154400.0</td>\n",
       "      <td>154200.0</td>\n",
       "      <td>...</td>\n",
       "      <td>558200</td>\n",
       "      <td>560800</td>\n",
       "      <td>562800</td>\n",
       "      <td>565600</td>\n",
       "      <td>569700</td>\n",
       "      <td>574000</td>\n",
       "      <td>577800</td>\n",
       "      <td>580600</td>\n",
       "      <td>583000</td>\n",
       "      <td>585100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>17426</td>\n",
       "      <td>Chicago</td>\n",
       "      <td>IL</td>\n",
       "      <td>Chicago</td>\n",
       "      <td>Cook</td>\n",
       "      <td>3</td>\n",
       "      <td>109700.0</td>\n",
       "      <td>109400.0</td>\n",
       "      <td>109300.0</td>\n",
       "      <td>109300.0</td>\n",
       "      <td>...</td>\n",
       "      <td>207800</td>\n",
       "      <td>206900</td>\n",
       "      <td>206200</td>\n",
       "      <td>205800</td>\n",
       "      <td>206200</td>\n",
       "      <td>207300</td>\n",
       "      <td>208200</td>\n",
       "      <td>209100</td>\n",
       "      <td>211000</td>\n",
       "      <td>213000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>13271</td>\n",
       "      <td>Philadelphia</td>\n",
       "      <td>PA</td>\n",
       "      <td>Philadelphia</td>\n",
       "      <td>Philadelphia</td>\n",
       "      <td>4</td>\n",
       "      <td>50000.0</td>\n",
       "      <td>49900.0</td>\n",
       "      <td>49600.0</td>\n",
       "      <td>49400.0</td>\n",
       "      <td>...</td>\n",
       "      <td>122300</td>\n",
       "      <td>121600</td>\n",
       "      <td>121800</td>\n",
       "      <td>123300</td>\n",
       "      <td>125200</td>\n",
       "      <td>126400</td>\n",
       "      <td>127000</td>\n",
       "      <td>127400</td>\n",
       "      <td>128300</td>\n",
       "      <td>129100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>40326</td>\n",
       "      <td>Phoenix</td>\n",
       "      <td>AZ</td>\n",
       "      <td>Phoenix</td>\n",
       "      <td>Maricopa</td>\n",
       "      <td>5</td>\n",
       "      <td>87200.0</td>\n",
       "      <td>87700.0</td>\n",
       "      <td>88200.0</td>\n",
       "      <td>88400.0</td>\n",
       "      <td>...</td>\n",
       "      <td>183800</td>\n",
       "      <td>185300</td>\n",
       "      <td>186600</td>\n",
       "      <td>188000</td>\n",
       "      <td>189100</td>\n",
       "      <td>190200</td>\n",
       "      <td>191300</td>\n",
       "      <td>192800</td>\n",
       "      <td>194500</td>\n",
       "      <td>195900</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 251 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   RegionID    RegionName State                           Metro    CountyName  \\\n",
       "0      6181      New York    NY                        New York        Queens   \n",
       "1     12447   Los Angeles    CA  Los Angeles-Long Beach-Anaheim   Los Angeles   \n",
       "2     17426       Chicago    IL                         Chicago          Cook   \n",
       "3     13271  Philadelphia    PA                    Philadelphia  Philadelphia   \n",
       "4     40326       Phoenix    AZ                         Phoenix      Maricopa   \n",
       "\n",
       "   SizeRank   1996-04   1996-05   1996-06   1996-07   ...     2015-11  \\\n",
       "0         1       NaN       NaN       NaN       NaN   ...      573600   \n",
       "1         2  155000.0  154600.0  154400.0  154200.0   ...      558200   \n",
       "2         3  109700.0  109400.0  109300.0  109300.0   ...      207800   \n",
       "3         4   50000.0   49900.0   49600.0   49400.0   ...      122300   \n",
       "4         5   87200.0   87700.0   88200.0   88400.0   ...      183800   \n",
       "\n",
       "   2015-12  2016-01  2016-02  2016-03  2016-04  2016-05  2016-06  2016-07  \\\n",
       "0   576200   578400   582200   588000   592200   592500   590200   588000   \n",
       "1   560800   562800   565600   569700   574000   577800   580600   583000   \n",
       "2   206900   206200   205800   206200   207300   208200   209100   211000   \n",
       "3   121600   121800   123300   125200   126400   127000   127400   128300   \n",
       "4   185300   186600   188000   189100   190200   191300   192800   194500   \n",
       "\n",
       "   2016-08  \n",
       "0   586400  \n",
       "1   585100  \n",
       "2   213000  \n",
       "3   129100  \n",
       "4   195900  \n",
       "\n",
       "[5 rows x 251 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "zillow = pd.read_csv('City_Zhvi_AllHomes.csv')\n",
    "zillow.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10730, 251)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "zillow.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So, after the first look at the data it seems we need the following cleaning:\n",
    "* The states are in short names, so we need to transform this.\n",
    "* It contains month by month prices from 1996, we only need data from 2001 so drop the unnecessary columns.\n",
    "* Use the state and region name for indexing.\n",
    "* Finally, convert month by month data to quarterly data.\n",
    "\n",
    "So, let's get to work!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def convert_housing_data_to_quarters():\n",
    "    '''Converts the housing data to quarters and returns it as mean housing price    \n",
    "    values in a dataframe. This  is a dataframe with\n",
    "    columns for 2000q1 through 2016q3, and have a multi-index\n",
    "    in the shape of [\"State\",\"RegionName\"].\n",
    "    \n",
    "    Note: Quarters are defined in the definition, they are\n",
    "    not arbitrary three month periods.\n",
    "        \n",
    "    '''\n",
    "    zillow = pd.read_csv('City_Zhvi_AllHomes.csv')\n",
    "    zillow['State'] = zillow['State'].map(states)\n",
    "    zillow.set_index(['State', 'RegionName'], inplace=True)\n",
    "    zillow = zillow.loc[:, '2000-01': ]\n",
    "    \n",
    "    new_columns = [str(x)+y for x in range(2000, 2017) for y in ['q1', 'q2', 'q3', 'q4']]\n",
    "    new_columns = new_columns[:-1] # drop the last quarter of 2016\n",
    "    \n",
    "    x = 0\n",
    "\n",
    "    for c in new_columns:\n",
    "        zillow[c] = zillow.iloc[:, x:x+3].mean(axis=1)\n",
    "        x = x+3\n",
    "    \n",
    "    zillow = zillow.loc[:, '2000q1':]\n",
    "    \n",
    "    \n",
    "    return zillow\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's take a look at the shape of the transformed zillow data. It should have the same number of rows as the original file that is 10730 and the number of columns should be 67 (4 quarters in 17 years minus the last  quarter in 2016)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(10730, 67)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "convert_housing_data_to_quarters().shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's have a peek at the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "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></th>\n",
       "      <th>2000q1</th>\n",
       "      <th>2000q2</th>\n",
       "      <th>2000q3</th>\n",
       "      <th>2000q4</th>\n",
       "      <th>2001q1</th>\n",
       "      <th>2001q2</th>\n",
       "      <th>2001q3</th>\n",
       "      <th>2001q4</th>\n",
       "      <th>2002q1</th>\n",
       "      <th>2002q2</th>\n",
       "      <th>...</th>\n",
       "      <th>2014q2</th>\n",
       "      <th>2014q3</th>\n",
       "      <th>2014q4</th>\n",
       "      <th>2015q1</th>\n",
       "      <th>2015q2</th>\n",
       "      <th>2015q3</th>\n",
       "      <th>2015q4</th>\n",
       "      <th>2016q1</th>\n",
       "      <th>2016q2</th>\n",
       "      <th>2016q3</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>State</th>\n",
       "      <th>RegionName</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>New York</th>\n",
       "      <th>New York</th>\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",
       "      <td>...</td>\n",
       "      <td>515466.666667</td>\n",
       "      <td>522800.000000</td>\n",
       "      <td>528066.666667</td>\n",
       "      <td>532266.666667</td>\n",
       "      <td>540800.000000</td>\n",
       "      <td>557200.000000</td>\n",
       "      <td>572833.333333</td>\n",
       "      <td>582866.666667</td>\n",
       "      <td>591633.333333</td>\n",
       "      <td>587200.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>California</th>\n",
       "      <th>Los Angeles</th>\n",
       "      <td>207066.666667</td>\n",
       "      <td>214466.666667</td>\n",
       "      <td>220966.666667</td>\n",
       "      <td>226166.666667</td>\n",
       "      <td>233000.000000</td>\n",
       "      <td>239100.000000</td>\n",
       "      <td>245066.666667</td>\n",
       "      <td>253033.333333</td>\n",
       "      <td>261966.666667</td>\n",
       "      <td>272700.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>498033.333333</td>\n",
       "      <td>509066.666667</td>\n",
       "      <td>518866.666667</td>\n",
       "      <td>528800.000000</td>\n",
       "      <td>538166.666667</td>\n",
       "      <td>547266.666667</td>\n",
       "      <td>557733.333333</td>\n",
       "      <td>566033.333333</td>\n",
       "      <td>577466.666667</td>\n",
       "      <td>458388.888889</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Illinois</th>\n",
       "      <th>Chicago</th>\n",
       "      <td>138400.000000</td>\n",
       "      <td>143633.333333</td>\n",
       "      <td>147866.666667</td>\n",
       "      <td>152133.333333</td>\n",
       "      <td>156933.333333</td>\n",
       "      <td>161800.000000</td>\n",
       "      <td>166400.000000</td>\n",
       "      <td>170433.333333</td>\n",
       "      <td>175500.000000</td>\n",
       "      <td>177566.666667</td>\n",
       "      <td>...</td>\n",
       "      <td>192633.333333</td>\n",
       "      <td>195766.666667</td>\n",
       "      <td>201266.666667</td>\n",
       "      <td>201066.666667</td>\n",
       "      <td>206033.333333</td>\n",
       "      <td>208300.000000</td>\n",
       "      <td>207900.000000</td>\n",
       "      <td>206066.666667</td>\n",
       "      <td>208200.000000</td>\n",
       "      <td>187466.666667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Pennsylvania</th>\n",
       "      <th>Philadelphia</th>\n",
       "      <td>53000.000000</td>\n",
       "      <td>53633.333333</td>\n",
       "      <td>54133.333333</td>\n",
       "      <td>54700.000000</td>\n",
       "      <td>55333.333333</td>\n",
       "      <td>55533.333333</td>\n",
       "      <td>56266.666667</td>\n",
       "      <td>57533.333333</td>\n",
       "      <td>59133.333333</td>\n",
       "      <td>60733.333333</td>\n",
       "      <td>...</td>\n",
       "      <td>113733.333333</td>\n",
       "      <td>115300.000000</td>\n",
       "      <td>115666.666667</td>\n",
       "      <td>116200.000000</td>\n",
       "      <td>117966.666667</td>\n",
       "      <td>121233.333333</td>\n",
       "      <td>122200.000000</td>\n",
       "      <td>123433.333333</td>\n",
       "      <td>126933.333333</td>\n",
       "      <td>103466.666667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Arizona</th>\n",
       "      <th>Phoenix</th>\n",
       "      <td>111833.333333</td>\n",
       "      <td>114366.666667</td>\n",
       "      <td>116000.000000</td>\n",
       "      <td>117400.000000</td>\n",
       "      <td>119600.000000</td>\n",
       "      <td>121566.666667</td>\n",
       "      <td>122700.000000</td>\n",
       "      <td>124300.000000</td>\n",
       "      <td>126533.333333</td>\n",
       "      <td>128366.666667</td>\n",
       "      <td>...</td>\n",
       "      <td>164266.666667</td>\n",
       "      <td>165366.666667</td>\n",
       "      <td>168500.000000</td>\n",
       "      <td>171533.333333</td>\n",
       "      <td>174166.666667</td>\n",
       "      <td>179066.666667</td>\n",
       "      <td>183833.333333</td>\n",
       "      <td>187900.000000</td>\n",
       "      <td>191433.333333</td>\n",
       "      <td>167411.111111</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 67 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                  2000q1         2000q2         2000q3  \\\n",
       "State        RegionName                                                  \n",
       "New York     New York                NaN            NaN            NaN   \n",
       "California   Los Angeles   207066.666667  214466.666667  220966.666667   \n",
       "Illinois     Chicago       138400.000000  143633.333333  147866.666667   \n",
       "Pennsylvania Philadelphia   53000.000000   53633.333333   54133.333333   \n",
       "Arizona      Phoenix       111833.333333  114366.666667  116000.000000   \n",
       "\n",
       "                                  2000q4         2001q1         2001q2  \\\n",
       "State        RegionName                                                  \n",
       "New York     New York                NaN            NaN            NaN   \n",
       "California   Los Angeles   226166.666667  233000.000000  239100.000000   \n",
       "Illinois     Chicago       152133.333333  156933.333333  161800.000000   \n",
       "Pennsylvania Philadelphia   54700.000000   55333.333333   55533.333333   \n",
       "Arizona      Phoenix       117400.000000  119600.000000  121566.666667   \n",
       "\n",
       "                                  2001q3         2001q4         2002q1  \\\n",
       "State        RegionName                                                  \n",
       "New York     New York                NaN            NaN            NaN   \n",
       "California   Los Angeles   245066.666667  253033.333333  261966.666667   \n",
       "Illinois     Chicago       166400.000000  170433.333333  175500.000000   \n",
       "Pennsylvania Philadelphia   56266.666667   57533.333333   59133.333333   \n",
       "Arizona      Phoenix       122700.000000  124300.000000  126533.333333   \n",
       "\n",
       "                                  2002q2      ...               2014q2  \\\n",
       "State        RegionName                       ...                        \n",
       "New York     New York                NaN      ...        515466.666667   \n",
       "California   Los Angeles   272700.000000      ...        498033.333333   \n",
       "Illinois     Chicago       177566.666667      ...        192633.333333   \n",
       "Pennsylvania Philadelphia   60733.333333      ...        113733.333333   \n",
       "Arizona      Phoenix       128366.666667      ...        164266.666667   \n",
       "\n",
       "                                  2014q3         2014q4         2015q1  \\\n",
       "State        RegionName                                                  \n",
       "New York     New York      522800.000000  528066.666667  532266.666667   \n",
       "California   Los Angeles   509066.666667  518866.666667  528800.000000   \n",
       "Illinois     Chicago       195766.666667  201266.666667  201066.666667   \n",
       "Pennsylvania Philadelphia  115300.000000  115666.666667  116200.000000   \n",
       "Arizona      Phoenix       165366.666667  168500.000000  171533.333333   \n",
       "\n",
       "                                  2015q2         2015q3         2015q4  \\\n",
       "State        RegionName                                                  \n",
       "New York     New York      540800.000000  557200.000000  572833.333333   \n",
       "California   Los Angeles   538166.666667  547266.666667  557733.333333   \n",
       "Illinois     Chicago       206033.333333  208300.000000  207900.000000   \n",
       "Pennsylvania Philadelphia  117966.666667  121233.333333  122200.000000   \n",
       "Arizona      Phoenix       174166.666667  179066.666667  183833.333333   \n",
       "\n",
       "                                  2016q1         2016q2         2016q3  \n",
       "State        RegionName                                                 \n",
       "New York     New York      582866.666667  591633.333333  587200.000000  \n",
       "California   Los Angeles   566033.333333  577466.666667  458388.888889  \n",
       "Illinois     Chicago       206066.666667  208200.000000  187466.666667  \n",
       "Pennsylvania Philadelphia  123433.333333  126933.333333  103466.666667  \n",
       "Arizona      Phoenix       187900.000000  191433.333333  167411.111111  \n",
       "\n",
       "[5 rows x 67 columns]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "convert_housing_data_to_quarters().head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So, the data is prepared as desired. Now, we will move into the final phase of the analysis.\n",
    "* We only need to look at the recession period, so we will narrow the columns to recession period.\n",
    "* We need a price ratio column to compare the prices between the start and bottom of the recession.\n",
    "* We need to split the data into university towns and non-university towns to run t-test on our hypothesis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def run_ttest():\n",
    "    '''First creates new data showing the decline or growth of housing prices\n",
    "    between the recession start and the recession bottom. Then runs a ttest\n",
    "    comparing the university town values to the non-university towns values, \n",
    "    return whether the alternative hypothesis (that the two groups are the same)\n",
    "    is true or not as well as the p-value of the confidence. \n",
    "    \n",
    "    Returns the tuple (different, p, lower) where different=True if the t-test is\n",
    "    True at a p<0.01 (we reject the null hypothesis), or different=False if \n",
    "    otherwise (we cannot reject the null hypothesis). The\n",
    "    value for lower should be either \"university town\" or \"non-university town\"\n",
    "    depending on which has a lower mean price ratio (which is equivilent to a\n",
    "    reduced market loss).'''\n",
    "\n",
    "    # get the start of recession\n",
    "    start = get_recession_start()\n",
    "    #get the bottom of the recession\n",
    "    bottom = get_recession_bottom()\n",
    "    # get the zillow housing data in desired format\n",
    "    housing_data = convert_housing_data_to_quarters()\n",
    "    # keep only the columns from recession starting point and bottom.\n",
    "    housing_data = housing_data.loc[:, start: bottom]\n",
    "    # compute price ratio between the recession points and add it as a column.\n",
    "    housing_data.reset_index(inplace=True)\n",
    "    housing_data['price_ratio'] = (housing_data[start] - housing_data[bottom]) / housing_data[start]\n",
    "    # get the university town list to split the data.\n",
    "    uni_towns = get_list_of_university_towns()  \n",
    "    uni_town_list = uni_towns['RegionName'].tolist()\n",
    "    # add a column to use as splitting condition\n",
    "    housing_data['isUniTown'] = housing_data.RegionName.apply(lambda x: x in uni_town_list)\n",
    "    #split the data into two separate dataframes and drop rows with missing values. The dropping step is needed to \n",
    "    #perform the t-test\n",
    "    uni_data = housing_data[housing_data.isUniTown].copy().dropna()\n",
    "    not_uni_data = housing_data[~housing_data.isUniTown].copy().dropna()\n",
    "    # get the p-value by applying t-test on these two dataframe columns.\n",
    "    p = ttest_ind(uni_data['price_ratio'], not_uni_data['price_ratio'])[1]\n",
    "    # this boolean value will tell us whether we can reject the null hyopthesis or not.\n",
    "    different = p < 0.01   \n",
    "    # this metric will tell us which type of town has the lower housing price ratio (mean) during the recession\n",
    "    lower = 'university town' if uni_data['price_ratio'].mean() < not_uni_data['price_ratio'].mean() else 'non-university town'\n",
    "    \n",
    "    return (different, p, lower)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, let's run the ttest to validate our hypothesis!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(True, 0.00036641601595526971, 'university town')"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "run_ttest()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**The p-value is well below the 0.05 threshold and so we can reject the null hypothesis and claim that housing prices in university towns are less affected by recession.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}