wrangle.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "43ee03f2",
   "metadata": {},
   "outputs": [],
   "source": [
    "from imports import *\n",
    "\n",
    "# set columnns display format\n",
    "pd.set_option('display.max_columns', None)\n",
    "# default pandas decimal number display format\n",
    "pd.options.display.float_format = '{:20,.3f}'.format"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5a9e6e84",
   "metadata": {},
   "source": [
    "# Data Acquisition"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "9a20d457",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_connection(db, user=env.user, host=env.host, password=env.password):\n",
    "    return f'mysql+pymysql://{user}:{password}@{host}/{db}'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "2ac42a03",
   "metadata": {},
   "outputs": [],
   "source": [
    "def new_zillow_data():\n",
    "    '''\n",
    "    This function reads the Zillow data from the mySQL database into a df.\n",
    "    '''\n",
    "    # Create SQL query.\n",
    "    sql_query = '''\n",
    "    SELECT \n",
    "    prop.*,\n",
    "    pred.logerror,\n",
    "    pred.transactiondate,\n",
    "    air.airconditioningdesc,\n",
    "    arch.architecturalstyledesc,\n",
    "    build.buildingclassdesc,\n",
    "    heat.heatingorsystemdesc,\n",
    "    landuse.propertylandusedesc,\n",
    "    story.storydesc,\n",
    "    construct.typeconstructiondesc\n",
    "FROM\n",
    "    properties_2017 prop\n",
    "        INNER JOIN\n",
    "    (SELECT \n",
    "        parcelid, logerror, MAX(transactiondate) AS transactiondate\n",
    "    FROM\n",
    "        predictions_2017\n",
    "    GROUP BY parcelid ,  logerror) pred USING (parcelid)\n",
    "        LEFT JOIN\n",
    "    airconditioningtype air USING (airconditioningtypeid)\n",
    "        LEFT JOIN\n",
    "    architecturalstyletype arch USING (architecturalstyletypeid)\n",
    "        LEFT JOIN\n",
    "    buildingclasstype build USING (buildingclasstypeid)\n",
    "        LEFT JOIN\n",
    "    heatingorsystemtype heat USING (heatingorsystemtypeid)\n",
    "        LEFT JOIN\n",
    "    propertylandusetype landuse USING (propertylandusetypeid)\n",
    "        LEFT JOIN\n",
    "    storytype story USING (storytypeid)\n",
    "        LEFT JOIN\n",
    "    typeconstructiontype construct USING (typeconstructiontypeid)\n",
    "WHERE prop.propertylandusetypeid = 261 AND\n",
    "        prop.latitude IS NOT NULL\n",
    "        AND prop.longitude IS NOT NULL\n",
    "        AND transactiondate <= '2017-12-31';\n",
    "    '''\n",
    "    # Read in DataFrame from Codeup db.\n",
    "    df = pd.read_sql(sql_query, get_connection('zillow'))\n",
    "    df = df.drop(columns='id')\n",
    "    return df\n",
    "\n",
    "def get_zillow_data():\n",
    "    '''\n",
    "    This function reads in zillow data from Zillow database, writes data to\n",
    "    a csv file if a local file does not exist, and returns a df.\n",
    "    '''\n",
    "    if os.path.isfile('zillow_data.csv'):\n",
    "        \n",
    "        # If csv file exists, read in data from csv file.\n",
    "        df = pd.read_csv('zillow_data.csv', index_col=0)\n",
    "    else:\n",
    "        \n",
    "        # Read fresh data from db into a DataFrame.\n",
    "        df = new_zillow_data()\n",
    "        \n",
    "        # Write DataFrame to a csv file.\n",
    "        df.to_csv('zillow_data.csv')\n",
    "        \n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "6c521750",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = get_zillow_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "e5735cd1",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(52438, 67)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "01f68a49",
   "metadata": {},
   "source": [
    "# Data Preparation"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2bc693b5",
   "metadata": {},
   "source": [
    "### 1. Missing Values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "5245a40e",
   "metadata": {},
   "outputs": [],
   "source": [
    "def handle_missing_values(df, prop_required_columns, prop_required_row):\n",
    "    threshold = int(round(prop_required_columns * len(df.index), 0))\n",
    "    df = df.dropna(axis=1, thresh=threshold) #1, or ‘columns’ : Drop columns which contain missing value\n",
    "    threshold = int(round(prop_required_row * len(df.columns), 0))\n",
    "    df = df.dropna(axis=0, thresh=threshold) #0, or ‘index’ : Drop rows which contain missing values.\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "80da16ba",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = handle_missing_values(df, 0.75, 0.75)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "56feeeb8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(52434, 28)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "2f5839e1",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.dropna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "fb621542",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "parcelid                        0\n",
       "bathroomcnt                     0\n",
       "bedroomcnt                      0\n",
       "calculatedbathnbr               0\n",
       "calculatedfinishedsquarefeet    0\n",
       "finishedsquarefeet12            0\n",
       "fips                            0\n",
       "fullbathcnt                     0\n",
       "latitude                        0\n",
       "longitude                       0\n",
       "lotsizesquarefeet               0\n",
       "propertycountylandusecode       0\n",
       "propertylandusetypeid           0\n",
       "rawcensustractandblock          0\n",
       "regionidcity                    0\n",
       "regionidcounty                  0\n",
       "regionidzip                     0\n",
       "roomcnt                         0\n",
       "yearbuilt                       0\n",
       "structuretaxvaluedollarcnt      0\n",
       "taxvaluedollarcnt               0\n",
       "assessmentyear                  0\n",
       "landtaxvaluedollarcnt           0\n",
       "taxamount                       0\n",
       "censustractandblock             0\n",
       "logerror                        0\n",
       "transactiondate                 0\n",
       "propertylandusedesc             0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "79a7e3d5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(50670, 28)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "5bb929c0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.03364229316855476"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(52434-50670)/52434"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2885085b",
   "metadata": {},
   "source": [
    "We feel comfortable dropping nulls because it's only 3.4% of the data."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "126455ad",
   "metadata": {},
   "source": [
    "### 2. Rename Columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "6ee69a75",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.rename(columns = {'bathroomcnt':'bathrooms', 'bedroomcnt':'bedrooms',\n",
    "                              'calculatedfinishedsquarefeet':'total_sqft', 'finishedsquarefeet12': 'living_sqft', 'fullbathcnt':'full_bath', 'lotsizesquarefeet':'lot_sqft', 'structuretaxvaluedollarcnt': 'structure_value', 'taxvaluedollarcnt':'assessed_value', 'landtaxvaluedollarcnt':'land_value'}, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "34457798",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['parcelid', 'bathrooms', 'bedrooms', 'calculatedbathnbr', 'total_sqft',\n",
       "       'living_sqft', 'fips', 'full_bath', 'latitude', 'longitude', 'lot_sqft',\n",
       "       'propertycountylandusecode', 'propertylandusetypeid',\n",
       "       'rawcensustractandblock', 'regionidcity', 'regionidcounty',\n",
       "       'regionidzip', 'roomcnt', 'yearbuilt', 'structure_value',\n",
       "       'assessed_value', 'assessmentyear', 'land_value', 'taxamount',\n",
       "       'censustractandblock', 'logerror', 'transactiondate',\n",
       "       'propertylandusedesc'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e11aeb3c",
   "metadata": {},
   "source": [
    "### 3. Data Mapping, Dummy Variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "ed16063c",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['age'] = 2017 - df['yearbuilt']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "021fd76e",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['county'] = df.fips.map({6037: 'Los Angeles', 6059: 'Orange', 6111: 'Ventura'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "b3de83c4",
   "metadata": {},
   "outputs": [],
   "source": [
    "dummy_df = pd.get_dummies(df[['county']], dummy_na=False, drop_first=False)\n",
    "df = pd.concat([df, dummy_df], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "b461be16",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['taxrate'] = df.taxamount/df.assessed_value*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "90383b56",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['transaction_month'] = df['transactiondate'].str.slice(5, 7)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "063e4907",
   "metadata": {},
   "source": [
    "### 4. Impute Latitude & Longitude"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "fd47b1b7",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.latitude = df.latitude/1000000"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "8d27998b",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.longitude = df.longitude/1000000"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "12d04102",
   "metadata": {},
   "source": [
    "### 5. Drop Unuseful Columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "659af311",
   "metadata": {},
   "outputs": [],
   "source": [
    "col = ['regionidzip','assessmentyear','yearbuilt','fips','propertycountylandusecode', 'propertylandusetypeid', 'rawcensustractandblock', 'regionidcounty', 'propertylandusedesc', 'transactiondate']\n",
    "df.drop(columns = col, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b8c13fba",
   "metadata": {},
   "source": [
    "### 6. Join Table to get zip code"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "aa6e7cf5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(50670, 25)"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "fc9385e9",
   "metadata": {},
   "outputs": [],
   "source": [
    "geo = pd.read_csv('address.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "c0fc64fe",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.merge(df, geo, on='parcelid', how='inner')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "2c3fa19c",
   "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>parcelid</th>\n",
       "      <th>bathrooms</th>\n",
       "      <th>bedrooms</th>\n",
       "      <th>calculatedbathnbr</th>\n",
       "      <th>total_sqft</th>\n",
       "      <th>living_sqft</th>\n",
       "      <th>full_bath</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>lot_sqft</th>\n",
       "      <th>regionidcity</th>\n",
       "      <th>roomcnt</th>\n",
       "      <th>structure_value</th>\n",
       "      <th>assessed_value</th>\n",
       "      <th>land_value</th>\n",
       "      <th>taxamount</th>\n",
       "      <th>censustractandblock</th>\n",
       "      <th>logerror</th>\n",
       "      <th>age</th>\n",
       "      <th>county</th>\n",
       "      <th>county_Los Angeles</th>\n",
       "      <th>county_Orange</th>\n",
       "      <th>county_Ventura</th>\n",
       "      <th>taxrate</th>\n",
       "      <th>transaction_month</th>\n",
       "      <th>zip_code</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>14297519</td>\n",
       "      <td>3.500</td>\n",
       "      <td>4.000</td>\n",
       "      <td>3.500</td>\n",
       "      <td>3,100.000</td>\n",
       "      <td>3,100.000</td>\n",
       "      <td>3.000</td>\n",
       "      <td>33.635</td>\n",
       "      <td>-117.869</td>\n",
       "      <td>4,506.000</td>\n",
       "      <td>53,571.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>485,713.000</td>\n",
       "      <td>1,023,282.000</td>\n",
       "      <td>537,569.000</td>\n",
       "      <td>11,013.720</td>\n",
       "      <td>60,590,630,072,012.000</td>\n",
       "      <td>0.026</td>\n",
       "      <td>19.000</td>\n",
       "      <td>Orange</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1.076</td>\n",
       "      <td>01</td>\n",
       "      <td>92,660.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>17052889</td>\n",
       "      <td>1.000</td>\n",
       "      <td>2.000</td>\n",
       "      <td>1.000</td>\n",
       "      <td>1,465.000</td>\n",
       "      <td>1,465.000</td>\n",
       "      <td>1.000</td>\n",
       "      <td>34.449</td>\n",
       "      <td>-119.282</td>\n",
       "      <td>12,647.000</td>\n",
       "      <td>13,091.000</td>\n",
       "      <td>5.000</td>\n",
       "      <td>88,000.000</td>\n",
       "      <td>464,000.000</td>\n",
       "      <td>376,000.000</td>\n",
       "      <td>5,672.480</td>\n",
       "      <td>61,110,010,023,006.000</td>\n",
       "      <td>0.056</td>\n",
       "      <td>50.000</td>\n",
       "      <td>Ventura</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.223</td>\n",
       "      <td>01</td>\n",
       "      <td>93,023.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>14186244</td>\n",
       "      <td>2.000</td>\n",
       "      <td>3.000</td>\n",
       "      <td>2.000</td>\n",
       "      <td>1,243.000</td>\n",
       "      <td>1,243.000</td>\n",
       "      <td>2.000</td>\n",
       "      <td>33.886</td>\n",
       "      <td>-117.823</td>\n",
       "      <td>8,432.000</td>\n",
       "      <td>21,412.000</td>\n",
       "      <td>6.000</td>\n",
       "      <td>85,289.000</td>\n",
       "      <td>564,778.000</td>\n",
       "      <td>479,489.000</td>\n",
       "      <td>6,488.300</td>\n",
       "      <td>60,590,218,022,012.000</td>\n",
       "      <td>0.005</td>\n",
       "      <td>55.000</td>\n",
       "      <td>Orange</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1.149</td>\n",
       "      <td>01</td>\n",
       "      <td>92,886.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>12177905</td>\n",
       "      <td>3.000</td>\n",
       "      <td>4.000</td>\n",
       "      <td>3.000</td>\n",
       "      <td>2,376.000</td>\n",
       "      <td>2,376.000</td>\n",
       "      <td>3.000</td>\n",
       "      <td>34.245</td>\n",
       "      <td>-118.241</td>\n",
       "      <td>13,038.000</td>\n",
       "      <td>396,551.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>108,918.000</td>\n",
       "      <td>145,143.000</td>\n",
       "      <td>36,225.000</td>\n",
       "      <td>1,777.510</td>\n",
       "      <td>60,373,001,001,006.000</td>\n",
       "      <td>-0.103</td>\n",
       "      <td>47.000</td>\n",
       "      <td>Los Angeles</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.225</td>\n",
       "      <td>01</td>\n",
       "      <td>91,214.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>12095076</td>\n",
       "      <td>3.000</td>\n",
       "      <td>4.000</td>\n",
       "      <td>3.000</td>\n",
       "      <td>2,962.000</td>\n",
       "      <td>2,962.000</td>\n",
       "      <td>3.000</td>\n",
       "      <td>34.145</td>\n",
       "      <td>-118.180</td>\n",
       "      <td>63,000.000</td>\n",
       "      <td>47,019.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>276,684.000</td>\n",
       "      <td>773,303.000</td>\n",
       "      <td>496,619.000</td>\n",
       "      <td>9,516.260</td>\n",
       "      <td>60,374,608,001,014.000</td>\n",
       "      <td>-0.001</td>\n",
       "      <td>67.000</td>\n",
       "      <td>Los Angeles</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.231</td>\n",
       "      <td>01</td>\n",
       "      <td>91,105.000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   parcelid            bathrooms             bedrooms    calculatedbathnbr  \\\n",
       "0  14297519                3.500                4.000                3.500   \n",
       "1  17052889                1.000                2.000                1.000   \n",
       "2  14186244                2.000                3.000                2.000   \n",
       "3  12177905                3.000                4.000                3.000   \n",
       "4  12095076                3.000                4.000                3.000   \n",
       "\n",
       "            total_sqft          living_sqft            full_bath  \\\n",
       "0            3,100.000            3,100.000                3.000   \n",
       "1            1,465.000            1,465.000                1.000   \n",
       "2            1,243.000            1,243.000                2.000   \n",
       "3            2,376.000            2,376.000                3.000   \n",
       "4            2,962.000            2,962.000                3.000   \n",
       "\n",
       "              latitude            longitude             lot_sqft  \\\n",
       "0               33.635             -117.869            4,506.000   \n",
       "1               34.449             -119.282           12,647.000   \n",
       "2               33.886             -117.823            8,432.000   \n",
       "3               34.245             -118.241           13,038.000   \n",
       "4               34.145             -118.180           63,000.000   \n",
       "\n",
       "          regionidcity              roomcnt      structure_value  \\\n",
       "0           53,571.000                0.000          485,713.000   \n",
       "1           13,091.000                5.000           88,000.000   \n",
       "2           21,412.000                6.000           85,289.000   \n",
       "3          396,551.000                0.000          108,918.000   \n",
       "4           47,019.000                0.000          276,684.000   \n",
       "\n",
       "        assessed_value           land_value            taxamount  \\\n",
       "0        1,023,282.000          537,569.000           11,013.720   \n",
       "1          464,000.000          376,000.000            5,672.480   \n",
       "2          564,778.000          479,489.000            6,488.300   \n",
       "3          145,143.000           36,225.000            1,777.510   \n",
       "4          773,303.000          496,619.000            9,516.260   \n",
       "\n",
       "     censustractandblock             logerror                  age  \\\n",
       "0 60,590,630,072,012.000                0.026               19.000   \n",
       "1 61,110,010,023,006.000                0.056               50.000   \n",
       "2 60,590,218,022,012.000                0.005               55.000   \n",
       "3 60,373,001,001,006.000               -0.103               47.000   \n",
       "4 60,374,608,001,014.000               -0.001               67.000   \n",
       "\n",
       "        county  county_Los Angeles  county_Orange  county_Ventura  \\\n",
       "0       Orange                   0              1               0   \n",
       "1      Ventura                   0              0               1   \n",
       "2       Orange                   0              1               0   \n",
       "3  Los Angeles                   1              0               0   \n",
       "4  Los Angeles                   1              0               0   \n",
       "\n",
       "               taxrate transaction_month             zip_code  \n",
       "0                1.076                01           92,660.000  \n",
       "1                1.223                01           93,023.000  \n",
       "2                1.149                01           92,886.000  \n",
       "3                1.225                01           91,214.000  \n",
       "4                1.231                01           91,105.000  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "8f88dfca",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.dropna()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5340e1bd",
   "metadata": {},
   "source": [
    "### 7. Data Types"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "33d90166",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "parcelid                 int64\n",
       "bathrooms              float64\n",
       "bedrooms               float64\n",
       "calculatedbathnbr      float64\n",
       "total_sqft             float64\n",
       "living_sqft            float64\n",
       "full_bath              float64\n",
       "latitude               float64\n",
       "longitude              float64\n",
       "lot_sqft               float64\n",
       "regionidcity           float64\n",
       "roomcnt                float64\n",
       "structure_value        float64\n",
       "assessed_value         float64\n",
       "land_value             float64\n",
       "taxamount              float64\n",
       "censustractandblock    float64\n",
       "logerror               float64\n",
       "age                    float64\n",
       "county                  object\n",
       "county_Los Angeles       uint8\n",
       "county_Orange            uint8\n",
       "county_Ventura           uint8\n",
       "taxrate                float64\n",
       "transaction_month       object\n",
       "zip_code               float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "4a2d7acc",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['zip_code'] = df['zip_code'].astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "6a1e82b0",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['age'] = df['age'].astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "becc6d19",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['transaction_month']=df['transaction_month'].astype(int)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ba84fdc1",
   "metadata": {},
   "source": [
    "### 8. Outliers"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "11aa6f97",
   "metadata": {},
   "source": [
    "#### Distribution of `bathroomcnt`, `bedroomcnt`, `calculatedbathnbr`, `calculatedfinishedsquarefeet`, `fullbathcnt`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "5eb829f3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x576 with 6 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (16,8))\n",
    "\n",
    "# List of columns\n",
    "cols = ['bathrooms', 'bedrooms', 'full_bath', 'roomcnt', 'total_sqft', 'lot_sqft']\n",
    "\n",
    "for i, col in enumerate(cols):\n",
    "    \n",
    "    # i starts at 0, but plot nos should start at 1\n",
    "    plot_number = i + 1 \n",
    "    \n",
    "    # Create subplot.\n",
    "    plt.subplot(1,6, plot_number)\n",
    "    \n",
    "    # Title with column name.\n",
    "    plt.title(col)\n",
    "    \n",
    "    # Display histogram for column.\n",
    "    sns.boxplot(data = df[col])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "a646bd98",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "6.0"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# bedrooms\n",
    "\n",
    "q1, q3 = df.bedrooms.quantile([.25, 0.75])\n",
    "iqr = q3 - q1\n",
    "k = 2\n",
    "upper_bound= q3 + k * iqr\n",
    "upper_bound"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "560efaf4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "6.0"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# bathrooms\n",
    "\n",
    "q1, q3 = df.bathrooms.quantile([.25, 0.75])\n",
    "iqr = q3 - q1\n",
    "k = 3\n",
    "upper_bound= q3 + k * iqr\n",
    "upper_bound"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "00c9b0aa",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "5340.0"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# total_sqft\n",
    "\n",
    "q1, q3 = df.total_sqft.quantile([.25, 0.75])\n",
    "iqr = q3 - q1\n",
    "k = 3\n",
    "upper_bound= q3 + k * iqr\n",
    "upper_bound"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "cd6988dc",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "40487.5"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# lot_sqft\n",
    "\n",
    "q1, q3 = df.lot_sqft.quantile([.25, 0.75])\n",
    "iqr = q3 - q1\n",
    "k = 10\n",
    "upper_bound= q3 + k * iqr\n",
    "upper_bound"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "55a521fb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "13510.625"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "q3+1.5*iqr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "e1bd4776",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(731, 25)"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df.lot_sqft>50000].shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4b604e27",
   "metadata": {},
   "source": [
    "#### Distribution of `assessed_value`, `structure_value`, `land_value`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "389e18ea",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x576 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize = (16,8))\n",
    "\n",
    "# List of columns\n",
    "cols = ['assessed_value', 'structure_value', 'land_value']\n",
    "\n",
    "for i, col in enumerate(cols):\n",
    "    \n",
    "    # i starts at 0, but plot nos should start at 1\n",
    "    plot_number = i + 1 \n",
    "    \n",
    "    # Create subplot.\n",
    "    plt.subplot(3,1, plot_number)\n",
    "    \n",
    "    # Title with column name.\n",
    "    plt.title(col)\n",
    "    \n",
    "    # Display histogram for column.\n",
    "    sns.histplot(data=df, x=col, kde=True, log_scale=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "2371f5a1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1071103.6"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# assessed_value\n",
    "\n",
    "q1, q3 = df.assessed_value.quantile([.25, 0.75])\n",
    "iqr = q3 - q1\n",
    "k = 1.1\n",
    "upper_bound= q3 + k * iqr\n",
    "upper_bound"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "15d48715",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = prepare.prep_zillow(df)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "25d31d20",
   "metadata": {},
   "source": [
    "### Create bins for zip_code"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "9c65945c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "93536    684\n",
       "93535    599\n",
       "93550    495\n",
       "93065    458\n",
       "93551    420\n",
       "        ... \n",
       "92970      1\n",
       "91204      1\n",
       "92397      1\n",
       "93013      1\n",
       "93563      1\n",
       "Name: zip_code, Length: 470, dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.zip_code.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "d5691a29",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.01714059492299396"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "overall_mean = df.logerror.mean()\n",
    "overall_mean"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "fdc87481",
   "metadata": {},
   "outputs": [],
   "source": [
    "logerror_zip = df.groupby('zip_code').logerror.mean().reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "b3a8668b",
   "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>zip_code</th>\n",
       "      <th>logerror</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>90001</td>\n",
       "      <td>0.004</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>90002</td>\n",
       "      <td>0.007</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>90003</td>\n",
       "      <td>-0.063</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>90004</td>\n",
       "      <td>-0.027</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>90005</td>\n",
       "      <td>0.015</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>465</th>\n",
       "      <td>93552</td>\n",
       "      <td>0.009</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>466</th>\n",
       "      <td>93553</td>\n",
       "      <td>-0.008</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>467</th>\n",
       "      <td>93563</td>\n",
       "      <td>0.146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>468</th>\n",
       "      <td>93591</td>\n",
       "      <td>0.047</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>469</th>\n",
       "      <td>93688</td>\n",
       "      <td>0.018</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>470 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     zip_code             logerror\n",
       "0       90001                0.004\n",
       "1       90002                0.007\n",
       "2       90003               -0.063\n",
       "3       90004               -0.027\n",
       "4       90005                0.015\n",
       "..        ...                  ...\n",
       "465     93552                0.009\n",
       "466     93553               -0.008\n",
       "467     93563                0.146\n",
       "468     93591                0.047\n",
       "469     93688                0.018\n",
       "\n",
       "[470 rows x 2 columns]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "logerror_zip"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "8c4b5f0b",
   "metadata": {},
   "outputs": [],
   "source": [
    "logerror_zip.to_csv('logeror_zip.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6f3d3af6",
   "metadata": {},
   "source": [
    "1 -> siginificantly higher\n",
    "2 -> significantly lower\n",
    "3 -> insignificantly higher\n",
    "4 -> insignificantly lower"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "a18996b3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.01714059492299396"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.logerror.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "2f0b0fa4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "90001\n",
      "4\n",
      "90002\n",
      "4\n",
      "90003\n",
      "2\n",
      "90004\n",
      "2\n",
      "90005\n",
      "4\n",
      "90006\n",
      "4\n",
      "90007\n",
      "4\n",
      "90008\n",
      "2\n",
      "90011\n",
      "4\n",
      "90012\n",
      "4\n",
      "90013\n",
      "4\n",
      "90014\n",
      "4\n",
      "90015\n",
      "4\n",
      "90016\n",
      "2\n",
      "90017\n",
      "4\n",
      "90018\n",
      "4\n",
      "90019\n",
      "3\n",
      "90020\n",
      "4\n",
      "90022\n",
      "3\n",
      "90023\n",
      "3\n",
      "90024\n",
      "2\n",
      "90025\n",
      "4\n",
      "90026\n",
      "2\n",
      "90027\n",
      "4\n",
      "90028\n",
      "4\n",
      "90029\n",
      "4\n",
      "90031\n",
      "2\n",
      "90032\n",
      "4\n",
      "90033\n",
      "4\n",
      "90034\n",
      "4\n",
      "90035\n",
      "4\n",
      "90036\n",
      "4\n",
      "90037\n",
      "4\n",
      "90038\n",
      "2\n",
      "90039\n",
      "2\n",
      "90040\n",
      "3\n",
      "90041\n",
      "2\n",
      "90042\n",
      "4\n",
      "90043\n",
      "3\n",
      "90044\n",
      "4\n",
      "90045\n",
      "2\n",
      "90046\n",
      "3\n",
      "90047\n",
      "4\n",
      "90048\n",
      "4\n",
      "90049\n",
      "3\n",
      "90053\n",
      "2\n",
      "90056\n",
      "4\n",
      "90057\n",
      "3\n",
      "90058\n",
      "3\n",
      "90059\n",
      "4\n",
      "90061\n",
      "3\n",
      "90062\n",
      "2\n",
      "90063\n",
      "4\n",
      "90064\n",
      "4\n",
      "90065\n",
      "4\n",
      "90066\n",
      "2\n",
      "90067\n",
      "3\n",
      "90068\n",
      "3\n",
      "90069\n",
      "2\n",
      "90071\n",
      "4\n",
      "90073\n",
      "4\n",
      "90077\n",
      "3\n",
      "90081\n",
      "4\n",
      "90089\n",
      "4\n",
      "90094\n",
      "4\n",
      "90095\n",
      "2\n",
      "90096\n",
      "4\n",
      "90201\n",
      "3\n",
      "90210\n",
      "3\n",
      "90211\n",
      "4\n",
      "90212\n",
      "4\n",
      "90220\n",
      "4\n",
      "90221\n",
      "4\n",
      "90222\n",
      "3\n",
      "90227\n",
      "4\n",
      "90230\n",
      "4\n",
      "90232\n",
      "2\n",
      "90240\n",
      "4\n",
      "90241\n",
      "3\n",
      "90242\n",
      "4\n",
      "90245\n",
      "3\n",
      "90247\n",
      "4\n",
      "90248\n",
      "4\n",
      "90249\n",
      "4\n",
      "90250\n",
      "4\n",
      "90254\n",
      "4\n",
      "90255\n",
      "4\n",
      "90260\n",
      "3\n",
      "90261\n",
      "4\n",
      "90262\n",
      "3\n",
      "90265\n",
      "1\n",
      "90266\n",
      "4\n",
      "90267\n",
      "4\n",
      "90270\n",
      "3\n",
      "90272\n",
      "2\n",
      "90274\n",
      "3\n",
      "90275\n",
      "3\n",
      "90277\n",
      "3\n",
      "90278\n",
      "2\n",
      "90280\n",
      "4\n",
      "90290\n",
      "3\n",
      "90291\n",
      "3\n",
      "90292\n",
      "3\n",
      "90293\n",
      "4\n",
      "90301\n",
      "4\n",
      "90302\n",
      "2\n",
      "90303\n",
      "4\n",
      "90304\n",
      "1\n",
      "90305\n",
      "2\n",
      "90306\n",
      "4\n",
      "90363\n",
      "2\n",
      "90401\n",
      "3\n",
      "90402\n",
      "4\n",
      "90403\n",
      "4\n",
      "90404\n",
      "3\n",
      "90405\n",
      "2\n",
      "90501\n",
      "4\n",
      "90502\n",
      "3\n",
      "90503\n",
      "3\n",
      "90504\n",
      "3\n",
      "90505\n",
      "3\n",
      "90506\n",
      "4\n",
      "90601\n",
      "3\n",
      "90602\n",
      "4\n",
      "90603\n",
      "2\n",
      "90604\n",
      "1\n",
      "90605\n",
      "4\n",
      "90606\n",
      "3\n",
      "90608\n",
      "4\n",
      "90620\n",
      "3\n",
      "90621\n",
      "3\n",
      "90623\n",
      "4\n",
      "90630\n",
      "4\n",
      "90631\n",
      "3\n",
      "90638\n",
      "2\n",
      "90639\n",
      "3\n",
      "90640\n",
      "3\n",
      "90650\n",
      "3\n",
      "90651\n",
      "3\n",
      "90660\n",
      "1\n",
      "90670\n",
      "3\n",
      "90680\n",
      "3\n",
      "90683\n",
      "4\n",
      "90701\n",
      "3\n",
      "90702\n",
      "4\n",
      "90703\n",
      "3\n",
      "90704\n",
      "3\n",
      "90706\n",
      "4\n",
      "90710\n",
      "4\n",
      "90712\n",
      "3\n",
      "90713\n",
      "3\n",
      "90715\n",
      "3\n",
      "90716\n",
      "3\n",
      "90717\n",
      "1\n",
      "90720\n",
      "1\n",
      "90723\n",
      "3\n",
      "90731\n",
      "4\n",
      "90732\n",
      "3\n",
      "90733\n",
      "4\n",
      "90740\n",
      "3\n",
      "90742\n",
      "4\n",
      "90744\n",
      "1\n",
      "90745\n",
      "4\n",
      "90746\n",
      "4\n",
      "90747\n",
      "4\n",
      "90755\n",
      "4\n",
      "90802\n",
      "3\n",
      "90803\n",
      "2\n",
      "90804\n",
      "4\n",
      "90805\n",
      "3\n",
      "90806\n",
      "4\n",
      "90807\n",
      "4\n",
      "90808\n",
      "3\n",
      "90810\n",
      "1\n",
      "90813\n",
      "3\n",
      "90814\n",
      "2\n",
      "90815\n",
      "4\n",
      "90822\n",
      "4\n",
      "90840\n",
      "3\n",
      "90843\n",
      "3\n",
      "90846\n",
      "4\n",
      "90851\n",
      "4\n",
      "91001\n",
      "4\n",
      "91006\n",
      "4\n",
      "91007\n",
      "3\n",
      "91008\n",
      "4\n",
      "91010\n",
      "3\n",
      "91011\n",
      "2\n",
      "91015\n",
      "4\n",
      "91016\n",
      "3\n",
      "91020\n",
      "4\n",
      "91024\n",
      "4\n",
      "91030\n",
      "2\n",
      "91040\n",
      "4\n",
      "91042\n",
      "4\n",
      "91046\n",
      "4\n",
      "91101\n",
      "4\n",
      "91103\n",
      "3\n",
      "91104\n",
      "2\n",
      "91105\n",
      "2\n",
      "91106\n",
      "2\n",
      "91107\n",
      "4\n",
      "91108\n",
      "2\n",
      "91109\n",
      "3\n",
      "91125\n",
      "4\n",
      "91184\n",
      "2\n",
      "91201\n",
      "4\n",
      "91202\n",
      "4\n",
      "91203\n",
      "1\n",
      "91205\n",
      "4\n",
      "91206\n",
      "1\n",
      "91207\n",
      "2\n",
      "91208\n",
      "2\n",
      "91214\n",
      "4\n",
      "91301\n",
      "3\n",
      "91302\n",
      "4\n",
      "91303\n",
      "1\n",
      "91304\n",
      "1\n",
      "91306\n",
      "4\n",
      "91307\n",
      "3\n",
      "91311\n",
      "3\n",
      "91312\n",
      "4\n",
      "91316\n",
      "3\n",
      "91320\n",
      "4\n",
      "91321\n",
      "3\n",
      "91324\n",
      "3\n",
      "91325\n",
      "3\n",
      "91326\n",
      "4\n",
      "91328\n",
      "3\n",
      "91330\n",
      "4\n",
      "91331\n",
      "3\n",
      "91335\n",
      "3\n",
      "91340\n",
      "4\n",
      "91342\n",
      "3\n",
      "91343\n",
      "4\n",
      "91344\n",
      "1\n",
      "91345\n",
      "4\n",
      "91350\n",
      "4\n",
      "91351\n",
      "3\n",
      "91352\n",
      "3\n",
      "91354\n",
      "4\n",
      "91355\n",
      "4\n",
      "91356\n",
      "3\n",
      "91360\n",
      "3\n",
      "91361\n",
      "2\n",
      "91362\n",
      "4\n",
      "91364\n",
      "1\n",
      "91365\n",
      "3\n",
      "91367\n",
      "4\n",
      "91371\n",
      "3\n",
      "91377\n",
      "4\n",
      "91384\n",
      "2\n",
      "91387\n",
      "3\n",
      "91390\n",
      "2\n",
      "91401\n",
      "4\n",
      "91402\n",
      "3\n",
      "91403\n",
      "4\n",
      "91405\n",
      "3\n",
      "91406\n",
      "2\n",
      "91411\n",
      "3\n",
      "91423\n",
      "4\n",
      "91436\n",
      "2\n",
      "91501\n",
      "3\n",
      "91502\n",
      "4\n",
      "91503\n",
      "4\n",
      "91504\n",
      "3\n",
      "91505\n",
      "4\n",
      "91506\n",
      "3\n",
      "91515\n",
      "4\n",
      "91522\n",
      "4\n",
      "91601\n",
      "3\n",
      "91602\n",
      "4\n",
      "91604\n",
      "4\n",
      "91605\n",
      "2\n",
      "91606\n",
      "4\n",
      "91607\n",
      "4\n",
      "91608\n",
      "3\n",
      "91702\n",
      "1\n",
      "91706\n",
      "1\n",
      "91711\n",
      "4\n",
      "91715\n",
      "3\n",
      "91722\n",
      "3\n",
      "91723\n",
      "3\n",
      "91724\n",
      "3\n",
      "91731\n",
      "4\n",
      "91732\n",
      "4\n",
      "91733\n",
      "3\n",
      "91740\n",
      "3\n",
      "91741\n",
      "3\n",
      "91742\n",
      "4\n",
      "91744\n",
      "3\n",
      "91745\n",
      "3\n",
      "91746\n",
      "3\n",
      "91748\n",
      "1\n",
      "91750\n",
      "3\n",
      "91754\n",
      "4\n",
      "91755\n",
      "4\n",
      "91763\n",
      "4\n",
      "91765\n",
      "3\n",
      "91766\n",
      "1\n",
      "91767\n",
      "4\n",
      "91768\n",
      "1\n",
      "91770\n",
      "3\n",
      "91773\n",
      "3\n",
      "91775\n",
      "3\n",
      "91776\n",
      "4\n",
      "91778\n",
      "3\n",
      "91780\n",
      "2\n",
      "91789\n",
      "3\n",
      "91790\n",
      "3\n",
      "91791\n",
      "3\n",
      "91792\n",
      "2\n",
      "91798\n",
      "2\n",
      "91801\n",
      "3\n",
      "91803\n",
      "3\n",
      "92406\n",
      "3\n",
      "92520\n",
      "3\n",
      "92602\n",
      "3\n",
      "92603\n",
      "3\n",
      "92604\n",
      "4\n",
      "92606\n",
      "4\n",
      "92607\n",
      "2\n",
      "92610\n",
      "2\n",
      "92612\n",
      "4\n",
      "92614\n",
      "2\n",
      "92615\n",
      "3\n",
      "92617\n",
      "3\n",
      "92618\n",
      "3\n",
      "92620\n",
      "3\n",
      "92624\n",
      "4\n",
      "92625\n",
      "3\n",
      "92626\n",
      "3\n",
      "92627\n",
      "3\n",
      "92629\n",
      "4\n",
      "92630\n",
      "3\n",
      "92637\n",
      "4\n",
      "92644\n",
      "3\n",
      "92646\n",
      "3\n",
      "92647\n",
      "3\n",
      "92648\n",
      "2\n",
      "92649\n",
      "3\n",
      "92651\n",
      "1\n",
      "92653\n",
      "3\n",
      "92655\n",
      "3\n",
      "92656\n",
      "4\n",
      "92657\n",
      "3\n",
      "92658\n",
      "3\n",
      "92660\n",
      "3\n",
      "92661\n",
      "4\n",
      "92662\n",
      "4\n",
      "92663\n",
      "1\n",
      "92666\n",
      "3\n",
      "92667\n",
      "4\n",
      "92669\n",
      "3\n",
      "92672\n",
      "3\n",
      "92673\n",
      "4\n",
      "92674\n",
      "3\n",
      "92675\n",
      "1\n",
      "92676\n",
      "4\n",
      "92677\n",
      "3\n",
      "92678\n",
      "4\n",
      "92679\n",
      "1\n",
      "92683\n",
      "3\n",
      "92684\n",
      "4\n",
      "92688\n",
      "3\n",
      "92689\n",
      "4\n",
      "92690\n",
      "3\n",
      "92691\n",
      "3\n",
      "92692\n",
      "4\n",
      "92694\n",
      "4\n",
      "92701\n",
      "3\n",
      "92703\n",
      "3\n",
      "92704\n",
      "3\n",
      "92705\n",
      "4\n",
      "92706\n",
      "3\n",
      "92707\n",
      "3\n",
      "92708\n",
      "4\n",
      "92763\n",
      "4\n",
      "92780\n",
      "3\n",
      "92781\n",
      "3\n",
      "92782\n",
      "2\n",
      "92801\n",
      "3\n",
      "92802\n",
      "4\n",
      "92804\n",
      "3\n",
      "92805\n",
      "4\n",
      "92806\n",
      "4\n",
      "92807\n",
      "3\n",
      "92808\n",
      "3\n",
      "92817\n",
      "3\n",
      "92821\n",
      "3\n",
      "92823\n",
      "3\n",
      "92831\n",
      "2\n",
      "92832\n",
      "3\n",
      "92833\n",
      "4\n",
      "92835\n",
      "4\n",
      "92840\n",
      "3\n",
      "92841\n",
      "4\n",
      "92842\n",
      "2\n",
      "92843\n",
      "4\n",
      "92844\n",
      "4\n",
      "92845\n",
      "3\n",
      "92861\n",
      "3\n",
      "92863\n",
      "4\n",
      "92865\n",
      "4\n",
      "92866\n",
      "4\n",
      "92867\n",
      "4\n",
      "92868\n",
      "3\n",
      "92869\n",
      "3\n",
      "92870\n",
      "1\n",
      "92885\n",
      "3\n",
      "92886\n",
      "3\n",
      "92887\n",
      "3\n",
      "93001\n",
      "4\n",
      "93002\n",
      "4\n",
      "93003\n",
      "4\n",
      "93004\n",
      "4\n",
      "93009\n",
      "4\n",
      "93010\n",
      "4\n",
      "93012\n",
      "4\n",
      "93015\n",
      "4\n",
      "93016\n",
      "3\n",
      "93021\n",
      "2\n",
      "93022\n",
      "4\n",
      "93023\n",
      "3\n",
      "93030\n",
      "4\n",
      "93031\n",
      "4\n",
      "93033\n",
      "3\n",
      "93035\n",
      "3\n",
      "93036\n",
      "3\n",
      "93040\n",
      "4\n",
      "93041\n",
      "3\n",
      "93043\n",
      "4\n",
      "93060\n",
      "3\n",
      "93061\n",
      "4\n",
      "93062\n",
      "2\n",
      "93063\n",
      "4\n",
      "93064\n",
      "3\n",
      "93065\n",
      "3\n",
      "93066\n",
      "3\n",
      "93510\n",
      "3\n",
      "93532\n",
      "3\n",
      "93534\n",
      "4\n",
      "93535\n",
      "2\n",
      "93536\n",
      "2\n",
      "93539\n",
      "3\n",
      "93543\n",
      "3\n",
      "93544\n",
      "1\n",
      "93550\n",
      "3\n",
      "93551\n",
      "4\n",
      "93552\n",
      "4\n",
      "93553\n",
      "4\n",
      "93591\n",
      "3\n",
      "93688\n",
      "3\n"
     ]
    }
   ],
   "source": [
    "for row in range(0,470):\n",
    "    alpha = 0.05\n",
    "    sample = df[df.zip_code ==logerror_zip['zip_code'][row]].logerror\n",
    "    t, p = stats.ttest_1samp(sample, overall_mean)\n",
    "    if p/2 < alpha:\n",
    "        print(logerror_zip['zip_code'][row])\n",
    "        if t>0:\n",
    "            print('1')\n",
    "        else:\n",
    "            print('2')\n",
    "    if p/2 >= alpha:\n",
    "        print(logerror_zip['zip_code'][row])\n",
    "        if t>0:\n",
    "            print('3')\n",
    "        else:\n",
    "            print('4')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "39ae965a",
   "metadata": {},
   "outputs": [
    {
     "ename": "KeyError",
     "evalue": "'cat'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
      "File \u001b[0;32m/usr/local/anaconda3/lib/python3.9/site-packages/pandas/core/indexes/base.py:3621\u001b[0m, in \u001b[0;36mIndex.get_loc\u001b[0;34m(self, key, method, tolerance)\u001b[0m\n\u001b[1;32m   3620\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m-> 3621\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_engine\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mget_loc\u001b[49m\u001b[43m(\u001b[49m\u001b[43mcasted_key\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   3622\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m err:\n",
      "File \u001b[0;32m/usr/local/anaconda3/lib/python3.9/site-packages/pandas/_libs/index.pyx:136\u001b[0m, in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[0;34m()\u001b[0m\n",
      "File \u001b[0;32m/usr/local/anaconda3/lib/python3.9/site-packages/pandas/_libs/index.pyx:163\u001b[0m, in \u001b[0;36mpandas._libs.index.IndexEngine.get_loc\u001b[0;34m()\u001b[0m\n",
      "File \u001b[0;32mpandas/_libs/hashtable_class_helper.pxi:5198\u001b[0m, in \u001b[0;36mpandas._libs.hashtable.PyObjectHashTable.get_item\u001b[0;34m()\u001b[0m\n",
      "File \u001b[0;32mpandas/_libs/hashtable_class_helper.pxi:5206\u001b[0m, in \u001b[0;36mpandas._libs.hashtable.PyObjectHashTable.get_item\u001b[0;34m()\u001b[0m\n",
      "\u001b[0;31mKeyError\u001b[0m: 'cat'",
      "\nThe above exception was the direct cause of the following exception:\n",
      "\u001b[0;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
      "Input \u001b[0;32mIn [50]\u001b[0m, in \u001b[0;36m<cell line: 2>\u001b[0;34m()\u001b[0m\n\u001b[1;32m     20\u001b[0m         logerror_zip[\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mcat\u001b[39m\u001b[38;5;124m'\u001b[39m][row] \u001b[38;5;241m==\u001b[39m \u001b[38;5;124m'\u001b[39m\u001b[38;5;124minsiginificantly high\u001b[39m\u001b[38;5;124m'\u001b[39m\n\u001b[1;32m     21\u001b[0m     \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m---> 22\u001b[0m         \u001b[43mlogerror_zip\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mcat\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m]\u001b[49m[row] \u001b[38;5;241m==\u001b[39m \u001b[38;5;124m'\u001b[39m\u001b[38;5;124minsiginificantly low\u001b[39m\u001b[38;5;124m'\u001b[39m\n\u001b[1;32m     23\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m     24\u001b[0m     logerror_zip[\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mcat\u001b[39m\u001b[38;5;124m'\u001b[39m][row] \u001b[38;5;241m==\u001b[39m \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mnone\u001b[39m\u001b[38;5;124m'\u001b[39m\n",
      "File \u001b[0;32m/usr/local/anaconda3/lib/python3.9/site-packages/pandas/core/frame.py:3505\u001b[0m, in \u001b[0;36mDataFrame.__getitem__\u001b[0;34m(self, key)\u001b[0m\n\u001b[1;32m   3503\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mcolumns\u001b[38;5;241m.\u001b[39mnlevels \u001b[38;5;241m>\u001b[39m \u001b[38;5;241m1\u001b[39m:\n\u001b[1;32m   3504\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_getitem_multilevel(key)\n\u001b[0;32m-> 3505\u001b[0m indexer \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mcolumns\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mget_loc\u001b[49m\u001b[43m(\u001b[49m\u001b[43mkey\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   3506\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m is_integer(indexer):\n\u001b[1;32m   3507\u001b[0m     indexer \u001b[38;5;241m=\u001b[39m [indexer]\n",
      "File \u001b[0;32m/usr/local/anaconda3/lib/python3.9/site-packages/pandas/core/indexes/base.py:3623\u001b[0m, in \u001b[0;36mIndex.get_loc\u001b[0;34m(self, key, method, tolerance)\u001b[0m\n\u001b[1;32m   3621\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_engine\u001b[38;5;241m.\u001b[39mget_loc(casted_key)\n\u001b[1;32m   3622\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m err:\n\u001b[0;32m-> 3623\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m(key) \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01merr\u001b[39;00m\n\u001b[1;32m   3624\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mTypeError\u001b[39;00m:\n\u001b[1;32m   3625\u001b[0m     \u001b[38;5;66;03m# If we have a listlike key, _check_indexing_error will raise\u001b[39;00m\n\u001b[1;32m   3626\u001b[0m     \u001b[38;5;66;03m#  InvalidIndexError. Otherwise we fall through and re-raise\u001b[39;00m\n\u001b[1;32m   3627\u001b[0m     \u001b[38;5;66;03m#  the TypeError.\u001b[39;00m\n\u001b[1;32m   3628\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_check_indexing_error(key)\n",
      "\u001b[0;31mKeyError\u001b[0m: 'cat'"
     ]
    }
   ],
   "source": [
    "row = 0\n",
    "while row<=469:\n",
    "    alpha = 0.05\n",
    "    sample = df[df.zip_code ==logerror_zip['zip_code'][row]].logerror\n",
    "    overall_mean = df.logerror.mean()\n",
    "    \n",
    "    significantly_high = []\n",
    "    significantly_low = []\n",
    "    insignificantly_high = []\n",
    "    insignificantly_low = []\n",
    "    \n",
    "    t, p = stats.ttest_1samp(sample, overall_mean)\n",
    "    if p/2 < alpha:\n",
    "        if t>0:\n",
    "            logerror_zip['cat'][row] == 'siginificantly high'\n",
    "        else:\n",
    "            logerror_zip['cat'][row] == 'siginificantly low'\n",
    "    if p/2 >= alpha:\n",
    "        if t>0:\n",
    "            logerror_zip['cat'][row] == 'insiginificantly high'\n",
    "        else:\n",
    "            logerror_zip['cat'][row] == 'insiginificantly low'\n",
    "    else:\n",
    "        logerror_zip['cat'][row] == 'none'\n",
    "    row+=1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "4215128e",
   "metadata": {},
   "outputs": [],
   "source": [
    "zip_error = pd.read_csv('logeror_zip.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "e8eae1ca",
   "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>zip_code</th>\n",
       "      <th>group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>90001</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>90002</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>90003</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>90004</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>90005</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>465</th>\n",
       "      <td>93552</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>466</th>\n",
       "      <td>93553</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>467</th>\n",
       "      <td>93563</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>468</th>\n",
       "      <td>93591</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>469</th>\n",
       "      <td>93688</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>470 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     zip_code  group\n",
       "0       90001      4\n",
       "1       90002      4\n",
       "2       90003      2\n",
       "3       90004      2\n",
       "4       90005      4\n",
       "..        ...    ...\n",
       "465     93552      4\n",
       "466     93553      4\n",
       "467     93563      1\n",
       "468     93591      3\n",
       "469     93688      3\n",
       "\n",
       "[470 rows x 2 columns]"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "zip_error"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "6997d22e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(45130, 23)"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "a90f5923",
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.merge(df, zip_error, on='zip_code', how='left')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "960ac754",
   "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>parcelid</th>\n",
       "      <th>bathrooms</th>\n",
       "      <th>bedrooms</th>\n",
       "      <th>total_sqft</th>\n",
       "      <th>living_sqft</th>\n",
       "      <th>full_bath</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>lot_sqft</th>\n",
       "      <th>roomcnt</th>\n",
       "      <th>structure_value</th>\n",
       "      <th>assessed_value</th>\n",
       "      <th>land_value</th>\n",
       "      <th>taxamount</th>\n",
       "      <th>logerror</th>\n",
       "      <th>age</th>\n",
       "      <th>taxrate</th>\n",
       "      <th>county</th>\n",
       "      <th>transaction_month</th>\n",
       "      <th>county_Los Angeles</th>\n",
       "      <th>county_Orange</th>\n",
       "      <th>county_Ventura</th>\n",
       "      <th>zip_code</th>\n",
       "      <th>group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>14297519</td>\n",
       "      <td>3.500</td>\n",
       "      <td>4.000</td>\n",
       "      <td>3,100.000</td>\n",
       "      <td>3,100.000</td>\n",
       "      <td>3.000</td>\n",
       "      <td>33.635</td>\n",
       "      <td>-117.869</td>\n",
       "      <td>4,506.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>485,713.000</td>\n",
       "      <td>1,023,282.000</td>\n",
       "      <td>537,569.000</td>\n",
       "      <td>11,013.720</td>\n",
       "      <td>0.026</td>\n",
       "      <td>19</td>\n",
       "      <td>1.076</td>\n",
       "      <td>Orange</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>92660</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>17052889</td>\n",
       "      <td>1.000</td>\n",
       "      <td>2.000</td>\n",
       "      <td>1,465.000</td>\n",
       "      <td>1,465.000</td>\n",
       "      <td>1.000</td>\n",
       "      <td>34.449</td>\n",
       "      <td>-119.282</td>\n",
       "      <td>12,647.000</td>\n",
       "      <td>5.000</td>\n",
       "      <td>88,000.000</td>\n",
       "      <td>464,000.000</td>\n",
       "      <td>376,000.000</td>\n",
       "      <td>5,672.480</td>\n",
       "      <td>0.056</td>\n",
       "      <td>50</td>\n",
       "      <td>1.223</td>\n",
       "      <td>Ventura</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>93023</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>14186244</td>\n",
       "      <td>2.000</td>\n",
       "      <td>3.000</td>\n",
       "      <td>1,243.000</td>\n",
       "      <td>1,243.000</td>\n",
       "      <td>2.000</td>\n",
       "      <td>33.886</td>\n",
       "      <td>-117.823</td>\n",
       "      <td>8,432.000</td>\n",
       "      <td>6.000</td>\n",
       "      <td>85,289.000</td>\n",
       "      <td>564,778.000</td>\n",
       "      <td>479,489.000</td>\n",
       "      <td>6,488.300</td>\n",
       "      <td>0.005</td>\n",
       "      <td>55</td>\n",
       "      <td>1.149</td>\n",
       "      <td>Orange</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>92886</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>12177905</td>\n",
       "      <td>3.000</td>\n",
       "      <td>4.000</td>\n",
       "      <td>2,376.000</td>\n",
       "      <td>2,376.000</td>\n",
       "      <td>3.000</td>\n",
       "      <td>34.245</td>\n",
       "      <td>-118.241</td>\n",
       "      <td>13,038.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>108,918.000</td>\n",
       "      <td>145,143.000</td>\n",
       "      <td>36,225.000</td>\n",
       "      <td>1,777.510</td>\n",
       "      <td>-0.103</td>\n",
       "      <td>47</td>\n",
       "      <td>1.225</td>\n",
       "      <td>Los Angeles</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>91214</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>12069064</td>\n",
       "      <td>1.000</td>\n",
       "      <td>2.000</td>\n",
       "      <td>738.000</td>\n",
       "      <td>738.000</td>\n",
       "      <td>1.000</td>\n",
       "      <td>34.149</td>\n",
       "      <td>-118.239</td>\n",
       "      <td>4,214.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>18,890.000</td>\n",
       "      <td>218,552.000</td>\n",
       "      <td>199,662.000</td>\n",
       "      <td>2,366.080</td>\n",
       "      <td>0.102</td>\n",
       "      <td>95</td>\n",
       "      <td>1.083</td>\n",
       "      <td>Los Angeles</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>91206</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>12790562</td>\n",
       "      <td>3.000</td>\n",
       "      <td>4.000</td>\n",
       "      <td>3,039.000</td>\n",
       "      <td>3,039.000</td>\n",
       "      <td>3.000</td>\n",
       "      <td>33.960</td>\n",
       "      <td>-118.007</td>\n",
       "      <td>20,028.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>177,527.000</td>\n",
       "      <td>220,583.000</td>\n",
       "      <td>43,056.000</td>\n",
       "      <td>3,104.190</td>\n",
       "      <td>-0.041</td>\n",
       "      <td>47</td>\n",
       "      <td>1.407</td>\n",
       "      <td>Los Angeles</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>90605</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>11104527</td>\n",
       "      <td>3.000</td>\n",
       "      <td>4.000</td>\n",
       "      <td>2,540.000</td>\n",
       "      <td>2,540.000</td>\n",
       "      <td>3.000</td>\n",
       "      <td>34.435</td>\n",
       "      <td>-118.464</td>\n",
       "      <td>10,384.000</td>\n",
       "      <td>0.000</td>\n",
       "      <td>271,715.000</td>\n",
       "      <td>430,108.000</td>\n",
       "      <td>158,393.000</td>\n",
       "      <td>6,103.360</td>\n",
       "      <td>0.006</td>\n",
       "      <td>18</td>\n",
       "      <td>1.419</td>\n",
       "      <td>Los Angeles</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>91351</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>13944538</td>\n",
       "      <td>2.500</td>\n",
       "      <td>3.000</td>\n",
       "      <td>1,340.000</td>\n",
       "      <td>1,340.000</td>\n",
       "      <td>2.000</td>\n",
       "      <td>33.771</td>\n",
       "      <td>-117.983</td>\n",
       "      <td>1,199.000</td>\n",
       "      <td>6.000</td>\n",
       "      <td>96,617.000</td>\n",
       "      <td>319,668.000</td>\n",
       "      <td>223,051.000</td>\n",
       "      <td>4,078.080</td>\n",
       "      <td>0.046</td>\n",
       "      <td>37</td>\n",
       "      <td>1.276</td>\n",
       "      <td>Orange</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>92844</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>17110996</td>\n",
       "      <td>2.500</td>\n",
       "      <td>3.000</td>\n",
       "      <td>1,371.000</td>\n",
       "      <td>1,371.000</td>\n",
       "      <td>2.000</td>\n",
       "      <td>34.238</td>\n",
       "      <td>-119.171</td>\n",
       "      <td>3,445.000</td>\n",
       "      <td>5.000</td>\n",
       "      <td>99,026.000</td>\n",
       "      <td>198,054.000</td>\n",
       "      <td>99,028.000</td>\n",
       "      <td>2,204.840</td>\n",
       "      <td>0.009</td>\n",
       "      <td>13</td>\n",
       "      <td>1.113</td>\n",
       "      <td>Ventura</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>93036</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>14375300</td>\n",
       "      <td>2.000</td>\n",
       "      <td>4.000</td>\n",
       "      <td>1,677.000</td>\n",
       "      <td>1,677.000</td>\n",
       "      <td>2.000</td>\n",
       "      <td>33.643</td>\n",
       "      <td>-117.707</td>\n",
       "      <td>5,304.000</td>\n",
       "      <td>8.000</td>\n",
       "      <td>169,118.000</td>\n",
       "      <td>528,847.000</td>\n",
       "      <td>359,729.000</td>\n",
       "      <td>5,496.440</td>\n",
       "      <td>-0.022</td>\n",
       "      <td>36</td>\n",
       "      <td>1.039</td>\n",
       "      <td>Orange</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>92630</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   parcelid            bathrooms             bedrooms           total_sqft  \\\n",
       "0  14297519                3.500                4.000            3,100.000   \n",
       "1  17052889                1.000                2.000            1,465.000   \n",
       "2  14186244                2.000                3.000            1,243.000   \n",
       "3  12177905                3.000                4.000            2,376.000   \n",
       "4  12069064                1.000                2.000              738.000   \n",
       "5  12790562                3.000                4.000            3,039.000   \n",
       "6  11104527                3.000                4.000            2,540.000   \n",
       "7  13944538                2.500                3.000            1,340.000   \n",
       "8  17110996                2.500                3.000            1,371.000   \n",
       "9  14375300                2.000                4.000            1,677.000   \n",
       "\n",
       "           living_sqft            full_bath             latitude  \\\n",
       "0            3,100.000                3.000               33.635   \n",
       "1            1,465.000                1.000               34.449   \n",
       "2            1,243.000                2.000               33.886   \n",
       "3            2,376.000                3.000               34.245   \n",
       "4              738.000                1.000               34.149   \n",
       "5            3,039.000                3.000               33.960   \n",
       "6            2,540.000                3.000               34.435   \n",
       "7            1,340.000                2.000               33.771   \n",
       "8            1,371.000                2.000               34.238   \n",
       "9            1,677.000                2.000               33.643   \n",
       "\n",
       "             longitude             lot_sqft              roomcnt  \\\n",
       "0             -117.869            4,506.000                0.000   \n",
       "1             -119.282           12,647.000                5.000   \n",
       "2             -117.823            8,432.000                6.000   \n",
       "3             -118.241           13,038.000                0.000   \n",
       "4             -118.239            4,214.000                0.000   \n",
       "5             -118.007           20,028.000                0.000   \n",
       "6             -118.464           10,384.000                0.000   \n",
       "7             -117.983            1,199.000                6.000   \n",
       "8             -119.171            3,445.000                5.000   \n",
       "9             -117.707            5,304.000                8.000   \n",
       "\n",
       "       structure_value       assessed_value           land_value  \\\n",
       "0          485,713.000        1,023,282.000          537,569.000   \n",
       "1           88,000.000          464,000.000          376,000.000   \n",
       "2           85,289.000          564,778.000          479,489.000   \n",
       "3          108,918.000          145,143.000           36,225.000   \n",
       "4           18,890.000          218,552.000          199,662.000   \n",
       "5          177,527.000          220,583.000           43,056.000   \n",
       "6          271,715.000          430,108.000          158,393.000   \n",
       "7           96,617.000          319,668.000          223,051.000   \n",
       "8           99,026.000          198,054.000           99,028.000   \n",
       "9          169,118.000          528,847.000          359,729.000   \n",
       "\n",
       "             taxamount             logerror  age              taxrate  \\\n",
       "0           11,013.720                0.026   19                1.076   \n",
       "1            5,672.480                0.056   50                1.223   \n",
       "2            6,488.300                0.005   55                1.149   \n",
       "3            1,777.510               -0.103   47                1.225   \n",
       "4            2,366.080                0.102   95                1.083   \n",
       "5            3,104.190               -0.041   47                1.407   \n",
       "6            6,103.360                0.006   18                1.419   \n",
       "7            4,078.080                0.046   37                1.276   \n",
       "8            2,204.840                0.009   13                1.113   \n",
       "9            5,496.440               -0.022   36                1.039   \n",
       "\n",
       "        county  transaction_month  county_Los Angeles  county_Orange  \\\n",
       "0       Orange                  1                   0              1   \n",
       "1      Ventura                  1                   0              0   \n",
       "2       Orange                  1                   0              1   \n",
       "3  Los Angeles                  1                   1              0   \n",
       "4  Los Angeles                  1                   1              0   \n",
       "5  Los Angeles                  1                   1              0   \n",
       "6  Los Angeles                  1                   1              0   \n",
       "7       Orange                  1                   0              1   \n",
       "8      Ventura                  1                   0              0   \n",
       "9       Orange                  1                   0              1   \n",
       "\n",
       "   county_Ventura  zip_code  group  \n",
       "0               0     92660      3  \n",
       "1               1     93023      3  \n",
       "2               0     92886      3  \n",
       "3               0     91214      4  \n",
       "4               0     91206      1  \n",
       "5               0     90605      4  \n",
       "6               0     91351      3  \n",
       "7               0     92844      4  \n",
       "8               1     93036      3  \n",
       "9               0     92630      3  "
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "71010fa9",
   "metadata": {},
   "source": [
    "### 9. Split Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ce38d8d4",
   "metadata": {},
   "outputs": [],
   "source": [
    "def split(df):\n",
    "    '''\n",
    "    This function drops the customer_id column and then splits a dataframe into \n",
    "    train, validate, and test in order to explore the data and to create and validate models. \n",
    "    It takes in a dataframe and contains an integer for setting a seed for replication. \n",
    "    Test is 20% of the original dataset. The remaining 80% of the dataset is \n",
    "    divided between valiidate and train, with validate being .30*.80= 24% of \n",
    "    the original dataset, and train being .70*.80= 56% of the original dataset. \n",
    "    The function returns, train, validate and test dataframes. \n",
    "    '''\n",
    "    train, test = train_test_split(df, test_size = .2, random_state=123)   \n",
    "    train, validate = train_test_split(train, test_size=.3, random_state=123)\n",
    "    \n",
    "    return train, validate, test"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}