{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Python Hello World Regression, A Look at scikit-learn "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this notebook we will do a basic data analysis in python.  \n",
    "\n",
    "Our goal is to see how the price of a used car depends on characteristics of the car (features.)  \n",
    "\n",
    "\n",
    "We will:\n",
    "\n",
    "* read in the data.  \n",
    "* transform and plot the data.  \n",
    "* fit a few simple regression models, getting fits, predictions and standard inference.\n",
    "* learn some basics of the key machine learning python package scikit-learn.\n",
    "* learn how to use categorical variables with dummies (one hot encoding).\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import Needed Libraries   \n",
    "\n",
    "We need to import numpy, pandas, and matplot.pyplot (as np, pd, and plt).  \n",
    "numpy gives as vector/matrix/array operations,  \n",
    "pandas gives us \"data frames\" data structures,  \n",
    "and matplot.pyplot give us graphics.  \n",
    "\n",
    "We also import LinearRegression from sklearn.linear_model to run the multiple regression.  \n",
    "\n",
    "We also import statsmodels.api (as sm) to get inference and summaries (e.g. R-squared, t-stats, p-values) for multiple regression.  \n",
    "\n",
    "We will use the OneHotEncoder to include a categorical variable in a linear regression model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "from sklearn.linear_model import LinearRegression\n",
    "import statsmodels.api as sm\n",
    "\n",
    "from sklearn.preprocessing import OneHotEncoder"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Read in the Data and Have a quick look at it\n",
    "\n",
    "\n",
    "First we will read in the data from the file susedcars.csv on Rob's data page on bitbucket.  \n",
    "\n",
    "You could also read directly from a local file.  \n",
    "Use %pwd, %cd, %ls to see your working directory, change your working directory,and list files in your working\n",
    "directory\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'/home/robert/do/webpage/python'"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "%pwd"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The commands preceded by % are \"magic commands\" that are part of ipython. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's read  the data directly from bitbucket.  \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "*** the type of cd is:\n",
      "<class 'pandas.core.frame.DataFrame'>\n"
     ]
    }
   ],
   "source": [
    "cd = pd.read_csv(\"https://bitbucket.org/remcc/rob-data-sets/downloads/susedcars.csv\")\n",
    "print(\"*** the type of cd is:\")\n",
    "print(type(cd))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The \"type\" in python is fantastic !!\n",
    "\n",
    "Now that we know it is a DataFrame, let's get some more basic information."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "***number number of rows and columns is:  (1000, 7)\n",
      "***the column names are:\n",
      "['price' 'trim' 'isOneOwner' 'mileage' 'year' 'color' 'displacement']\n"
     ]
    }
   ],
   "source": [
    "print(\"***number number of rows and columns is: \",cd.shape)\n",
    "print(\"***the column names are:\")\n",
    "print(cd.columns.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "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>price</th>\n",
       "      <th>trim</th>\n",
       "      <th>isOneOwner</th>\n",
       "      <th>mileage</th>\n",
       "      <th>year</th>\n",
       "      <th>color</th>\n",
       "      <th>displacement</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>43995</td>\n",
       "      <td>550</td>\n",
       "      <td>f</td>\n",
       "      <td>36858.0</td>\n",
       "      <td>2008</td>\n",
       "      <td>Silver</td>\n",
       "      <td>5.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>44995</td>\n",
       "      <td>550</td>\n",
       "      <td>f</td>\n",
       "      <td>46883.0</td>\n",
       "      <td>2012</td>\n",
       "      <td>Black</td>\n",
       "      <td>4.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>25999</td>\n",
       "      <td>550</td>\n",
       "      <td>f</td>\n",
       "      <td>108759.0</td>\n",
       "      <td>2007</td>\n",
       "      <td>White</td>\n",
       "      <td>5.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>33880</td>\n",
       "      <td>550</td>\n",
       "      <td>f</td>\n",
       "      <td>35187.0</td>\n",
       "      <td>2007</td>\n",
       "      <td>Black</td>\n",
       "      <td>5.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>34895</td>\n",
       "      <td>550</td>\n",
       "      <td>f</td>\n",
       "      <td>48153.0</td>\n",
       "      <td>2007</td>\n",
       "      <td>Black</td>\n",
       "      <td>5.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   price trim isOneOwner   mileage  year   color displacement\n",
       "0  43995  550          f   36858.0  2008  Silver          5.5\n",
       "1  44995  550          f   46883.0  2012   Black          4.6\n",
       "2  25999  550          f  108759.0  2007   White          5.5\n",
       "3  33880  550          f   35187.0  2007   Black          5.5\n",
       "4  34895  550          f   48153.0  2007   Black          5.5"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cd.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Each of the 1,000 rows corresponds to a used car.  \n",
    "\n",
    "price is what the car sold for.   \n",
    "\n",
    "The other variables are *features* describing the car.   \n",
    "Our goal is to relate the price to the other features.\n",
    "\n",
    "Note the notation for head.  \n",
    "We call the head *method* on the object referred to by cd."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can pull one column (variable) out of the data frame by name."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     36858.0\n",
       "1     46883.0\n",
       "2    108759.0\n",
       "3     35187.0\n",
       "4     48153.0\n",
       "Name: mileage, dtype: float64"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp = cd['mileage']  # pull out the variable mileage\n",
    "temp.head()  # print out the mileage of the first few cars"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " The feature mileage is a numeric variable with units miles. We can summarize it using the usual descriptive summaries:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count      1000.000000\n",
       "mean      73652.408000\n",
       "std       42887.422189\n",
       "min        1997.000000\n",
       "25%       40132.750000\n",
       "50%       67919.500000\n",
       "75%      100138.250000\n",
       "max      255419.000000\n",
       "Name: mileage, dtype: float64"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cd['mileage'].describe()   # summary statistics of variable mileage"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The feature color is a categorical variable.  Each car is in one of the color categories.  We can't summarize a categorical variable the same way that we\n",
    "summarize a numeric variable.  There is no \"average\" color.  To summarize a\n",
    "categorical variable we simply count how many observations are in each category."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0    Silver\n",
      "1     Black\n",
      "2     White\n",
      "3     Black\n",
      "4     Black\n",
      "Name: color, dtype: object\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "color\n",
       "Black     415\n",
       "other     227\n",
       "Silver    213\n",
       "White     145\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(cd['color'].head())  # colors of first 5 cars\n",
    "cd['color'].value_counts() # how many cars have each color"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Use iloc to subset a data frame\n",
    "\n",
    "You can also use integers to pick off rows and columns using **iloc**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['price', 'trim', 'isOneOwner', 'mileage', 'year', 'color',\n",
       "       'displacement'], dtype=object)"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cd.columns.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "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>year</th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2008</td>\n",
       "      <td>43995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2012</td>\n",
       "      <td>44995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2007</td>\n",
       "      <td>25999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2007</td>\n",
       "      <td>33880</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2007</td>\n",
       "      <td>34895</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   year  price\n",
       "0  2008  43995\n",
       "1  2012  44995\n",
       "2  2007  25999\n",
       "3  2007  33880\n",
       "4  2007  34895"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "XXdf = cd.iloc[:,[4,0]] #year and price\n",
    "XXdf.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "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>year</th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2008</td>\n",
       "      <td>43995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2012</td>\n",
       "      <td>44995</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2007</td>\n",
       "      <td>25999</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   year  price\n",
       "0  2008  43995\n",
       "1  2012  44995\n",
       "2  2007  25999"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cd.iloc[0:3,[4,0]] #pick off rows and columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Note that in python, indexing starts at 0.** \n",
    "\n",
    "The range of indices a to b gives from a to b-1, b is not included.\n",
    "\n",
    "For example:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1\n",
      "7\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[1, 2, 3]"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp = [1,2,3,7]\n",
    "print(temp[0]) # the first element\n",
    "print(temp[-1]) # the last elemnt\n",
    "temp[0:3] #the 0, 1, and 2 element of the list temp"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Pick off  y=price and x=(mileage,year) and scale"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's focus on the two numeric features mileage and year.   \n",
    "Our goal will be to see how price relates to mileage and year.  \n",
    "We will divide both price and mileage by 1,000 to make the results easier to understand.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    price  mileage  year\n",
      "0  43.995   36.858  2008\n",
      "1  44.995   46.883  2012\n",
      "2  25.999  108.759  2007\n",
      "3  33.880   35.187  2007\n",
      "4  34.895   48.153  2007\n"
     ]
    }
   ],
   "source": [
    "cd = cd[['price','mileage','year']]\n",
    "cd['price']  /= 1000\n",
    "cd['mileage']  /= 1000\n",
    "print(cd.head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "             price      mileage         year\n",
      "count  1000.000000  1000.000000  1000.000000\n",
      "mean     30.583318    73.652408  2006.939000\n",
      "std      18.411018    42.887422     4.194624\n",
      "min       0.995000     1.997000  1994.000000\n",
      "25%      12.995000    40.132750  2004.000000\n",
      "50%      29.800000    67.919500  2007.000000\n",
      "75%      43.992000   100.138250  2010.000000\n",
      "max      79.995000   255.419000  2013.000000\n"
     ]
    }
   ],
   "source": [
    "print(cd.describe()) #summarize each column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "            price   mileage      year\n",
      "price    1.000000 -0.815246  0.880537\n",
      "mileage -0.815246  1.000000 -0.744729\n",
      "year     0.880537 -0.744729  1.000000\n"
     ]
    }
   ],
   "source": [
    "print(cd.corr()) #compute the correlation between each column"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Remember, a correlation is between -1 and 1.  \n",
    "\n",
    "The closer the correlation is to 1, the stronger the linear relationship between the variables,\n",
    "with a positive slope.  \n",
    "\n",
    "The closer the correlation is to -1, the stronger the linear relationship between the variables,\n",
    "with a negative slope. \n",
    "\n",
    "So it looks like the bigger the mileage is, the lower the price of the car.  \n",
    "The bigger the year is, the higher the price of the car.  \n",
    "\n",
    "Makes sense!!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Y and x, Features  and Target\n",
    "\n",
    "We often use \"y\" to generically denote the variable we trying to predict and \"x\" to denote\n",
    "the variables we can use to predict y.  \n",
    "\n",
    "In our example y=price and x=(mileage,year).  \n",
    "\n",
    "x=(mileage,year) is the what we know about the car.  Given this knowledge, what is our\n",
    "guess for the price of the car.  \n",
    "\n",
    "As we have done above, x is also often called the *features*.  \n",
    "x is also often called the independent variables.\n",
    "\n",
    "Y (or y) is often called the dependent variable or the target."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Functions, methods, attributes, data structures\n",
    "\n",
    "Python has a few key data structures built in and then we have to know a few more.  \n",
    "    \n",
    "Built in ones are:\n",
    "* tuple\n",
    "* lists\n",
    "* dictionaries\n",
    "\n",
    "We also have to work with numpy ndarrays and pandas Series and DataFrames.\n",
    "\n",
    "As we can then call functions with objects as arguments and outputs.  \n",
    "\n",
    "Python is object orientated so we also need to be aware of methods and attributes of objects.\n",
    "\n",
    "A couple of examples:\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "#### tuple\n",
      "<class 'tuple'>\n",
      "1\n",
      "\n",
      "#### list\n",
      "<class 'list'>\n",
      "length of x is 3\n",
      "[1, 3.4, 'robert']\n",
      "\n",
      "#### dictionaries\n",
      "{'rob': 14, 'ira': 13}\n",
      "{'rob': 7, 'ira': 13}\n",
      "dict_keys(['rob', 'ira'])\n",
      "dict_values([7, 13])\n"
     ]
    }
   ],
   "source": [
    "## a tuple\n",
    "print('\\n#### tuple')\n",
    "x = (1,2) \n",
    "print(type(x))\n",
    "\n",
    "print(x[0])\n",
    "#x[0]= 7.4  # does not work, tuple is immutable\n",
    "# note y = (5,) not y=(5) for a typle with one element\n",
    "\n",
    "## a list\n",
    "print('\\n#### list')\n",
    "x = [1,3.4,'rob']\n",
    "print(type(x))\n",
    "print(f'length of x is {len(x)}')\n",
    "x[2] = 'robert'\n",
    "print(x)\n",
    "\n",
    "## a dictionaries\n",
    "print('\\n#### dictionaries')\n",
    "x = {'rob': 14, 'ira': 13}\n",
    "print(x)\n",
    "x['rob'] = 7  # you access the value from the key\n",
    "print(x)\n",
    "print(x.keys())\n",
    "print(x.values())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "#### ndarray\n",
      "[[12.  14.7]\n",
      " [13.2 55.8]\n",
      " [34.  77.8]]\n",
      "(3, 2)\n",
      "float64\n"
     ]
    }
   ],
   "source": [
    "## numpy ndarray\n",
    "print('\\n#### ndarray')\n",
    "x = np.array([[12,14.7],[13.2,55.8],[34,77.8]])\n",
    "print(x)\n",
    "print(x.shape)\n",
    "print(x.dtype)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'tuple'>\n",
      "<class 'builtin_function_or_method'>\n",
      "34.583333333333336\n",
      "34.583333333333336\n",
      "[19.73333333 49.43333333]\n",
      "[13.35 34.5  55.9 ]\n",
      "\n",
      "\n",
      "\n",
      "[12.  13.2 34. ]\n",
      "[12.  13.2]\n",
      "[14.7 77.8]\n"
     ]
    }
   ],
   "source": [
    "print(type(x.shape)) ## attribute\n",
    "print(type(x.mean)) ## method\n",
    "print(x.mean()) # method\n",
    "print(np.mean(x)) # function\n",
    "print(x.mean(axis=0)) #column means\n",
    "print(x.mean(axis=1)) # row means\n",
    "\n",
    "print('\\n\\n')\n",
    "print(x[:,0]) #first column\n",
    "print(x[0:2,0]) # first two rows, first column\n",
    "print(x[[0,2],1]) # first and third row, second column"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "#### DataFrame\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "      0     1\n",
      "0  12.0  14.7\n",
      "1  13.2  55.8\n",
      "2  34.0  77.8\n"
     ]
    }
   ],
   "source": [
    "## DataFrame\n",
    "print('\\n#### DataFrame')\n",
    "xdf = pd.DataFrame(x) #DataFrame from a numpy ndarray\n",
    "print(type(xdf))\n",
    "print(xdf)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "If you call the function dir on an object, you get a list of all the attributes and methods!!  \n",
    "The methods and attributes with __ are for \"internal use\"."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['__add__',\n",
       " '__class__',\n",
       " '__class_getitem__',\n",
       " '__contains__',\n",
       " '__delattr__',\n",
       " '__delitem__',\n",
       " '__dir__',\n",
       " '__doc__',\n",
       " '__eq__',\n",
       " '__format__',\n",
       " '__ge__',\n",
       " '__getattribute__',\n",
       " '__getitem__',\n",
       " '__gt__',\n",
       " '__hash__',\n",
       " '__iadd__',\n",
       " '__imul__',\n",
       " '__init__',\n",
       " '__init_subclass__',\n",
       " '__iter__',\n",
       " '__le__',\n",
       " '__len__',\n",
       " '__lt__',\n",
       " '__mul__',\n",
       " '__ne__',\n",
       " '__new__',\n",
       " '__reduce__',\n",
       " '__reduce_ex__',\n",
       " '__repr__',\n",
       " '__reversed__',\n",
       " '__rmul__',\n",
       " '__setattr__',\n",
       " '__setitem__',\n",
       " '__sizeof__',\n",
       " '__str__',\n",
       " '__subclasshook__',\n",
       " 'append',\n",
       " 'clear',\n",
       " 'copy',\n",
       " 'count',\n",
       " 'extend',\n",
       " 'index',\n",
       " 'insert',\n",
       " 'pop',\n",
       " 'remove',\n",
       " 'reverse',\n",
       " 'sort']"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x=[1,2,3]\n",
    "dir(x)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "jp-MarkdownHeadingCollapsed": true
   },
   "source": [
    "## Help\n",
    "\n",
    "If you want help on anything with name junk,  just do junk? <return>.  \n",
    "This works on variables, modules, and just about anything!!   \n",
    "\n",
    "Tab completion and wild cards are very powerfull.  \n",
    "Try:  \n",
    "```x = np.arange(0,5)```  \n",
    "```x?```   \n",
    "```x.<TAB>```  \n",
    "```x.m<TAB>```  \n",
    "```import sta<TAB>```   \n",
    "```from sklearn.linear_model import L<TAB>```    \n",
    "```np.<TAB>```  \n",
    "```np.*mean*?```   "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Get y=price and X=(mileage,year) as numpy ndarrays  \n",
    "\n",
    "Let's get a numpy array X whose 2 columns are the explanatory features *mileage* and *year*.  \n",
    "\n",
    "Let's also get a numpy array with just the target variable y = *price*.\n",
    "\n",
    "You don't have to go to numpy, you could stick with pandas objects, but I like numpy for this."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "*** type of X is <class 'numpy.ndarray'>\n",
      "(1000, 2)\n",
      "[[  36.858 2008.   ]\n",
      " [  46.883 2012.   ]\n",
      " [ 108.759 2007.   ]\n",
      " [  35.187 2007.   ]]\n",
      "length of y is 1000\n",
      "[43.995 44.995 25.999 33.88 ]\n"
     ]
    }
   ],
   "source": [
    "X = cd[['mileage','year']].to_numpy()  #mileage and year columns as a numpy array\n",
    "print(\"*** type of X is\",type(X))\n",
    "print(X.shape) #number of rows and columns\n",
    "print(X[0:4,:]) #first 4 rows\n",
    "y = cd['price'].values #price as a numpy vector\n",
    "print(f'length of y is {len(y)}')\n",
    "print(y[:4]) #implicit start at 0"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Plot y vs each x, histogram of y  \n",
    "\n",
    "Now let's plot year vs. price.  \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'year vs. price')"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(X[:,1],y)\n",
    "plt.xlabel(\"year\")\n",
    "plt.ylabel(\"price\")\n",
    "plt.title(\"year vs. price\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And mileage vs. price.  \n",
    "Let's change the plot symbol, the size of the plotted symbol and the color of the plotted symbol.  \n",
    "We will also change the size of the figure."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(X[:,0],y,s=20,c=\"red\",marker='o')\n",
    "plt.xlabel(\"mileage\")\n",
    "plt.ylabel(\"price\")\n",
    "plt.title(\"mileage vs. price\")\n",
    "figure = plt.gcf()\n",
    "figure.set_size_inches(8,6)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's look at the histogram of price.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'price')"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "p = sns.histplot(y,bins=20)\n",
    "p.set_xlabel('price')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Clearly, price is related to both year and mileage.  \n",
    "Clearly, the relationship is not linear !!!  \n",
    "\n",
    "What we really want to **learn** is the joint relationship betwee *price* and the pair of variables\n",
    "(*mileage*,*year*) !!!  \n",
    "\n",
    "The modern statistical tools of *Machine Learning* enables us to learn the relationships\n",
    "from data without making strong assumptions.\n",
    "\n",
    "In the expression:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$$\n",
    "price = f(mileage,year)\n",
    "$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "we would like to know the function $f$."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### plot with pandas\n",
    "\n",
    "You can do a lot of the plotting directly in pandas."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='year', ylabel='price'>"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cd.plot.scatter(2,0,c=\"blue\") #access columns 0 and 2 = mileage and price"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Axes: xlabel='mileage', ylabel='price'>"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "cd.plot.scatter('mileage','price',c=\"red\",s=.5) # access columns using names"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A nice seaborn plot:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.PairGrid at 0x73611ac4f550>"
      ]
     },
     "execution_count": 91,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x600 with 12 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.pairplot(cd,diag_kind='kde',height=2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Regression with just mileage\n",
    "\n",
    "Our goal is to relate y=price to x=(mileage,price).  \n",
    "Let's start simple and just use mileage.  \n",
    "\n",
    "Our *simple linear regression model* is\n",
    "\n",
    "$$\n",
    "price = \\beta_0 + \\beta_1 \\, mileage + \\epsilon\n",
    "$$\n",
    "\n",
    "$\\epsilon$ represents the part of price we cannot learn from mileage.  \n",
    "\n",
    "Let's get a numpy array with just one column for mileage.  \n",
    "I'll reshape it so that it is a two dimensional array instead of a vector.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1000,)\n",
      "(1000, 1)\n",
      "[[ 36.858]\n",
      " [ 46.883]\n",
      " [108.759]\n",
      " [ 35.187]]\n"
     ]
    }
   ],
   "source": [
    "X1 = X[:,0] # first column of X is mileage\n",
    "print(X1.shape)\n",
    "X1 = X1.reshape((X.shape[0],1))\n",
    "print(X1.shape)\n",
    "print(X1[:4])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ok, now let's fit our simple linear regression model using mileage.  \n",
    "\n",
    "We will use LinearRegression from sklearn.  \n",
    "It will be a two step process.  \n",
    "We first create the model object and then we call the fit method with the training data. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model Slope:     [-0.34997452]\n",
      "Model Intercept: 56.359784475930795\n"
     ]
    }
   ],
   "source": [
    "lmmod1 = LinearRegression(fit_intercept=True) #model object\n",
    "lmmod1.fit(X1,y) # (X1,y) is the training data\n",
    "print(\"Model Slope:    \",lmmod1.coef_)\n",
    "print(\"Model Intercept:\",lmmod1.intercept_)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Our fitted model is\n",
    "$$\n",
    "price = 56.36 - .35 \\, mileage + \\epsilon\n",
    "$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's plot the training data with the fitted line.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(-5.0, 82.0)"
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "yhat = lmmod1.intercept_ + lmmod1.coef_ * X1\n",
    "plt.scatter(X1,y,c='blue',s=.7,label='data')\n",
    "plt.plot(X1,yhat,c='red',label='linear fit')\n",
    "plt.legend()\n",
    "plt.xlabel('x=mileage'); plt.ylabel('y=price')\n",
    "plt.title(\"mileage vs price with linear fit\")\n",
    "plt.ylim((-5,82))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pretty bad!!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Run The Regression of y=price on X=(mileage,year)  \n",
    "\n",
    "Now let's run a linear regression of *price* on *mileage* and *year*.  \n",
    "\n",
    "Our model is:  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "$$\n",
    "price = \\beta_0 + \\beta_1 mileage + \\beta_2 year + \\epsilon\n",
    "$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This model assumes a linear relationship.  \n",
    "\n",
    "*We already suspect this may be a bad idea* !!!  \n",
    "\n",
    "Let's go ahead and *fit* the model.  \n",
    "\n",
    "Fitting the model to data will give us estimates of the \n",
    "parameters $(\\beta_0,\\beta_1,\\beta_2)$.  \n",
    "\n",
    "The error term $\\epsilon$ represents the part of price we cannot\n",
    "know from (*mileage*,*year*).  \n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model Slopes:     [-0.1537219   2.69434954]\n",
      "Model Intercept: -5365.489872256993\n"
     ]
    }
   ],
   "source": [
    "lmmod = LinearRegression(fit_intercept=True)\n",
    "lmmod.fit(X,y) # (X,y) is the training data\n",
    "print(\"Model Slopes:    \",lmmod.coef_)\n",
    "print(\"Model Intercept:\",lmmod.intercept_)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Note that there does not seem to be a simple regression summary in sklearn.\n",
    "Maybe that is a good thing !!!!.\n",
    "\n",
    "So, the fitted relationship is  \n",
    "$$\n",
    "price = -5365.49  - 0.154 \\, mileage + 2.7 \\, year\n",
    "$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If we use just one feature (mileage above), we can visualize the regression fit in a very simple way.\n",
    "Even with just two x's, it is much harder to see what is going on and with many x's it is not so easy at all but  very interesting."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Looking at the LinearRegression object lmmod\n",
    "\n",
    "Let's have a quick look at the lmmod object."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "LinearRegression()\n",
      "{'copy_X': True, 'fit_intercept': True, 'n_jobs': None, 'positive': False}\n"
     ]
    }
   ],
   "source": [
    "print(lmmod)\n",
    "print(lmmod.get_params())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Above we see the basic attributed you can set when using LinearRegression.  \n",
    "Some are obvious, like fit_intercept controls whether or not an intercept is included in the regression.  \n",
    "For others you would try (i) ?lmmod (ii) Read the sklearn documentation (iii) google it, (iv) read a book."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['__abstractmethods__',\n",
       " '__annotations__',\n",
       " '__class__',\n",
       " '__delattr__',\n",
       " '__dict__',\n",
       " '__dir__',\n",
       " '__doc__',\n",
       " '__eq__',\n",
       " '__format__',\n",
       " '__ge__',\n",
       " '__getattribute__',\n",
       " '__getstate__',\n",
       " '__gt__',\n",
       " '__hash__',\n",
       " '__init__',\n",
       " '__init_subclass__',\n",
       " '__le__',\n",
       " '__lt__',\n",
       " '__module__',\n",
       " '__ne__',\n",
       " '__new__',\n",
       " '__reduce__',\n",
       " '__reduce_ex__',\n",
       " '__repr__',\n",
       " '__setattr__',\n",
       " '__setstate__',\n",
       " '__sizeof__',\n",
       " '__sklearn_clone__',\n",
       " '__str__',\n",
       " '__subclasshook__',\n",
       " '__weakref__',\n",
       " '_abc_impl',\n",
       " '_build_request_for_signature',\n",
       " '_check_feature_names',\n",
       " '_check_n_features',\n",
       " '_decision_function',\n",
       " '_estimator_type',\n",
       " '_get_default_requests',\n",
       " '_get_metadata_request',\n",
       " '_get_param_names',\n",
       " '_get_tags',\n",
       " '_more_tags',\n",
       " '_parameter_constraints',\n",
       " '_repr_html_',\n",
       " '_repr_html_inner',\n",
       " '_repr_mimebundle_',\n",
       " '_set_intercept',\n",
       " '_validate_data',\n",
       " '_validate_params',\n",
       " 'coef_',\n",
       " 'copy_X',\n",
       " 'fit',\n",
       " 'fit_intercept',\n",
       " 'get_metadata_routing',\n",
       " 'get_params',\n",
       " 'intercept_',\n",
       " 'n_features_in_',\n",
       " 'n_jobs',\n",
       " 'positive',\n",
       " 'predict',\n",
       " 'rank_',\n",
       " 'score',\n",
       " 'set_fit_request',\n",
       " 'set_params',\n",
       " 'set_score_request',\n",
       " 'singular_']"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dir(lmmod) #you can always find out a lot about an object with dir() !!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Some of the things in llmod are attributes (data structures) and some are methods (functions).\n",
    "Usually the members with trailing underscores (like coef_) are data members and the members without are methods."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'numpy.ndarray'>\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "method"
      ]
     },
     "execution_count": 98,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(type(lmmod.coef_))\n",
    "type(lmmod.set_params)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So, you could do ?lmmod.set_params at a python prompt."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Get and Plot the Fits \n",
    "\n",
    "Let's get the fitted values.  \n",
    "\n",
    "For each observation in our data set the fits are\n",
    "$$\n",
    "\\hat{price}_i = -5365.49  - 0.154 \\, mileage_i + 2.7 \\, year_i, \\;\\; i=1,2,\\ldots,n.\n",
    "$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "You can think of the fit as the predicted price give the values of *mileage* and *year* according\n",
    "to the model.  \n",
    "\n",
    "Notice that the coefficient for mileage is different from the one we got in the simple linear\n",
    "regression of price on mileage alone !!\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "the length of yhat is 1000\n",
      "the type of yhat is:\n",
      "<class 'numpy.ndarray'>\n"
     ]
    }
   ],
   "source": [
    "yhat = lmmod.predict(X)\n",
    "print(\"the length of yhat is\",len(yhat))\n",
    "print(\"the type of yhat is:\")\n",
    "print(type(yhat))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'yhat')"
      ]
     },
     "execution_count": 100,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(y,yhat,s=.8) \n",
    "plt.plot(y,y,c='red') #add the line\n",
    "plt.xlabel(\"y\"); plt.ylabel(\"yhat\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Clearly, it is really bad !!!**  \n",
    "\n",
    "Machine Learning will enable us to get it right fairly automatically.  \n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Predictions\n",
    "\n",
    "Let's get predictions for *x* not in our training data.  \n",
    "\n",
    "We will make a numpy array whose rows have the x values we want to predict at.  \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[  40. 2010.]\n",
      " [ 100. 2004.]]\n",
      "<class 'numpy.ndarray'>\n",
      "float64\n"
     ]
    }
   ],
   "source": [
    "Xp = np.array([[40,2010],[100,2004]],dtype=float)\n",
    "print(Xp)\n",
    "print(type(Xp))\n",
    "print(Xp.dtype)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So, the first car has 40 (thousand) miles on it and is a 2010, while\n",
    "the second car has 100 (thousand) miles on it and is a 2004.  \n",
    "Clearly, we expect the second car to sell for less!  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[44.00383414 18.61442272]\n"
     ]
    }
   ],
   "source": [
    "ypred = lmmod.predict(Xp)\n",
    "print(ypred)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So we predict (based on the linear model) that the first car will sell for\n",
    "44 (thou) and the second car will sell for 18.6."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's check the first one \"by hand\".  \n",
    "\n",
    "Model Slopes:     [-0.1537219   2.69434954]  \n",
    "Model Intercept: -5365.489872256993\n",
    "\n",
    "So the prediction for the first car in ```Xp``` should be:  \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "44.00457540000025"
      ]
     },
     "execution_count": 103,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "-5365.49 - .1537*40 + 2.69434954*2010"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "which is correct."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## In-sample/out of sample, training data, test data\n",
    "\n",
    "The data we used to \"fit\" our model, is called the *training data*.  \n",
    "When we look at predictions for observations in the training data\n",
    "(as we did for ```yhat```)\n",
    "we say we are looking at *in-sample* fits.  \n",
    "\n",
    "When we predict at observations not in the training data (as we did for ```ypred```), then we are\n",
    "predicting *out of sample*.  \n",
    "\n",
    "Out of sample prediction is the name of the game in predictive modelling or *supervised learning* !!!!\n",
    "\n",
    "In supervised learning, we *learn* the parameters of our model on *training data* and then predict on the \n",
    "out of sample *test data*.  A good model makes accurate predictions on the out of sample test data.   \n",
    "We predict the *target variable* y with the *features* x.  \n",
    "As noted above, in our example y = price x=(mileage,year).   \n",
    "In statistics y would be called the dependent variable and x would be called the independent variables.  \n",
    "\n",
    "There is also unsupervised learning in which the goal is to understand struture in data without neccessarily   \n",
    "have a target variable y that we want to predict.   \n",
    "\n",
    "For example, if I regress y=price on x=mileage, I am doing supervised learning.  \n",
    "If I compute the correlation between y and x that is unsupervised learning.  \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## scikit-learn\n",
    "\n",
    "Linear Regression is a basic model.  \n",
    "There are many modeling approaches in Machine Learning !!  \n",
    "scikit-learn has a nice general approach to working with models:\n",
    "\n",
    "    - a model will have a set of hyperparameters (e.g. lmmod.fit_intercept)  \n",
    "    - given the hyparameters, the model can learn from training data by calling the fit method (e.g lmmod.fit(X,y))  \n",
    "    - given a model has learned, it can make predictions (e.g. lmmod.predict(Xp))  \n",
    " \n",
    "All the predictive models in scikit-learn use the basic setup."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Standard Regression Output\n",
    "\n",
    "From our linear regression fit using sklearn, we got estimates for the parameters.  \n",
    "\n",
    "Often we want to know a lot more about the model fit.  \n",
    "\n",
    "In particular,  we might want to know the *standard errors* associated with the parameter\n",
    "estimates.  \n",
    "\n",
    "To get the usual *regression ouput* we can us the python package statsmodels, imported above as sm."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[1.00000e+00 3.68580e+01 2.00800e+03]\n",
      " [1.00000e+00 4.68830e+01 2.01200e+03]\n",
      " [1.00000e+00 1.08759e+02 2.00700e+03]]\n",
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                      y   R-squared:                       0.832\n",
      "Model:                            OLS   Adj. R-squared:                  0.832\n",
      "Method:                 Least Squares   F-statistic:                     2477.\n",
      "Date:                Sun, 11 Jan 2026   Prob (F-statistic):               0.00\n",
      "Time:                        10:44:14   Log-Likelihood:                -3438.1\n",
      "No. Observations:                1000   AIC:                             6882.\n",
      "Df Residuals:                     997   BIC:                             6897.\n",
      "Df Model:                           2                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "const      -5365.4899    171.567    -31.273      0.000   -5702.164   -5028.816\n",
      "x1            -0.1537      0.008    -18.435      0.000      -0.170      -0.137\n",
      "x2             2.6943      0.085     31.602      0.000       2.527       2.862\n",
      "==============================================================================\n",
      "Omnibus:                      171.937   Durbin-Watson:                   2.021\n",
      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              294.618\n",
      "Skew:                           1.076   Prob(JB):                     1.06e-64\n",
      "Kurtosis:                       4.562   Cond. No.                     1.44e+06\n",
      "==============================================================================\n",
      "\n",
      "Notes:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
      "[2] The condition number is large, 1.44e+06. This might indicate that there are\n",
      "strong multicollinearity or other numerical problems.\n"
     ]
    }
   ],
   "source": [
    "X = sm.add_constant(X) #appends 1 to beginning of each row for the intercept\n",
    "print(X[0:3,:]) # you can see the 1's\n",
    "results = sm.OLS(y, X).fit() #run the regression\n",
    "print(results.summary()) # print out the usual summaries"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Lot's of junk!!   \n",
    "\n",
    "In particular, the standard error associate with the estimate of the slope for *mileage* is .008.  \n",
    "\n",
    "The 95\\% confidence interval for $\\beta_1$, the *mileage* slope is estimate +/- 2 standard errors:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-0.1697, -0.1377])"
      ]
     },
     "execution_count": 105,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "-0.1537 + np.array([-2,2])*0.008"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Recall that $R^2$ is the square of the correlation between $y$ and $\\hat{y}$:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1.        , 0.91238967],\n",
       "       [0.91238967, 1.        ]])"
      ]
     },
     "execution_count": 106,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.corrcoef(y,yhat)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8324555121"
      ]
     },
     "execution_count": 107,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    ".91239**2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Which is the same as the ```R-squared``` in the regression output.\n",
    "\n",
    "Let's compare to the R-squared with just mileage.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1.        , 0.81524579],\n",
       "       [0.81524579, 1.        ]])"
      ]
     },
     "execution_count": 108,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yhat1 = lmmod1.predict(X1)\n",
    "np.corrcoef(y,yhat1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.664225"
      ]
     },
     "execution_count": 109,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    ".815**2"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Let's try the variable trim\n",
    "\n",
    "Does knowing the trim of a car help us predict the price ?  \n",
    "\n",
    "Let's try it.  \n",
    "How do we add the **categorical variable trim** to a linear regression model?  \n",
    "\n",
    "Let's pull the 1's back out of X.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1000, 2)\n",
      "[[  36.858 2008.   ]\n",
      " [  46.883 2012.   ]\n",
      " [ 108.759 2007.   ]\n",
      " [  35.187 2007.   ]\n",
      " [  48.153 2007.   ]]\n",
      "(1000,)\n",
      "[43.995 44.995 25.999 33.88  34.895]\n"
     ]
    }
   ],
   "source": [
    "Xo = X[:,1:]\n",
    "print(Xo.shape)\n",
    "print(Xo[:5,:])\n",
    "print(y.shape)\n",
    "print(y[:5])      "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we have to make the dummies for trim.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'numpy.ndarray'>\n",
      "(1000, 4)\n",
      "trim\n",
      "550      591\n",
      "430      143\n",
      "other    139\n",
      "500      127\n",
      "Name: count, dtype: int64\n",
      "[143. 127. 591. 139.]\n"
     ]
    }
   ],
   "source": [
    "## read in the data again to get trim\n",
    "cdn = pd.read_csv(\"https://bitbucket.org/remcc/rob-data-sets/downloads/susedcars.csv\")\n",
    "## create binary dummies (one hot encode) trim\n",
    "dent = OneHotEncoder(sparse_output=False)\n",
    "dums = dent.fit_transform(cdn[['trim']])\n",
    "## check dums\n",
    "print(type(dums))\n",
    "print(dums.shape)\n",
    "print(cdn['trim'].value_counts())\n",
    "print(dums.sum(axis=0))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So there are 4 different categories for trim.\n",
    "The first column of dums corresponds to trim = 430, the second to trim=500, the third to trim=550, and fourth to other."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0    550\n",
      "1    550\n",
      "2    550\n",
      "3    550\n",
      "4    550\n",
      "5    500\n",
      "Name: trim, dtype: object\n",
      "[[0. 0. 1. 0.]\n",
      " [0. 0. 1. 0.]\n",
      " [0. 0. 1. 0.]\n",
      " [0. 0. 1. 0.]\n",
      " [0. 0. 1. 0.]\n",
      " [0. 1. 0. 0.]]\n"
     ]
    }
   ],
   "source": [
    "print(cdn['trim'][:6])\n",
    "print(dums[:6,:])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "ok, now let's add three of the dummies to x and run the regression.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[3.68580e+01, 2.00800e+03, 0.00000e+00, 1.00000e+00, 0.00000e+00],\n",
       "       [4.68830e+01, 2.01200e+03, 0.00000e+00, 1.00000e+00, 0.00000e+00],\n",
       "       [1.08759e+02, 2.00700e+03, 0.00000e+00, 1.00000e+00, 0.00000e+00],\n",
       "       [3.51870e+01, 2.00700e+03, 0.00000e+00, 1.00000e+00, 0.00000e+00],\n",
       "       [4.81530e+01, 2.00700e+03, 0.00000e+00, 1.00000e+00, 0.00000e+00],\n",
       "       [1.21748e+02, 2.00200e+03, 1.00000e+00, 0.00000e+00, 0.00000e+00]])"
      ]
     },
     "execution_count": 113,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "XX = np.hstack([Xo,dums[:,1:]]) # drop the first column of dums\n",
    "XX[:6,:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "model slopes: [-0.1393757   2.65822306  2.42715984  4.77477184  9.58923747]\n",
      "model intercept: -5298.505897549599\n"
     ]
    }
   ],
   "source": [
    "lmmodc = LinearRegression(fit_intercept=True)\n",
    "lmmodc.fit(XX,y)\n",
    "print('model slopes:',lmmodc.coef_)\n",
    "print('model intercept:',lmmodc.intercept_)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now let's plot the fits."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'yhat with trim')"
      ]
     },
     "execution_count": 115,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAj4AAAGwCAYAAACpYG+ZAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy81sbWrAAAACXBIWXMAAA9hAAAPYQGoP6dpAABn60lEQVR4nO3deVxUVf8H8M+wCiojuAAuKGCKC+VWgluYlJFZPtmuPvTYrpZGm2bqU2aULZaWmtWvLCuzekzL0IxccsHKFfc1MWRJ0UEFAWfu74/rvc6dmQtzYYbZPu/Xi9fImTt3zp2x7tfz/Z5zdIIgCCAiIiLyAX6u7gARERFRfWHgQ0RERD6DgQ8RERH5DAY+RERE5DMY+BAREZHPYOBDREREPoOBDxEREfmMAFd3wJ2YTCacPHkSjRs3hk6nc3V3iIiIyA6CIODcuXNo2bIl/PyqH9Nh4GPm5MmTaNOmjau7QURERLVw4sQJtG7dutpjGPiYady4MQDxgwsLC3Nxb4iIiMgepaWlaNOmjXwfrw4DHzNSeissLIyBDxERkYexp0yFxc1ERETkMxj4EBERkc9g4ENEREQ+g4EPERER+QwGPkREROQzGPgQERGRz2DgQ0RERD6DgQ8RERH5DAY+RERE5DMY+BAREZHP8JjAx2g0YsqUKYiNjUVISAji4+Mxffp0CIIgHyMIAqZOnYro6GiEhIQgNTUVhw4dcmGviYiIyJ14TODz+uuvY968eXjvvfewb98+vP7665g5cybmzJkjHzNz5kzMnj0b8+fPx5YtW9CwYUMMHjwYFy9edGHPiYiIyF3oBPMhEzd26623IjIyEh9//LHcNnz4cISEhGDRokUQBAEtW7bE008/jWeeeQYAYDAYEBkZiU8//RT33ntvje9RWloKvV4Pg8HATUqJiIgcraoK0OmAAMfuka7l/u0xIz59+vRBdnY2Dh48CADYuXMnNmzYgLS0NADAsWPHUFhYiNTUVPk1er0evXv3xubNm22es6KiAqWlpYofIiIicrDSUqBhQyAoCLj7bpd2xWMCn4kTJ+Lee+9FQkICAgMD0b17d0yYMAEjRowAABQWFgIAIiMjFa+LjIyUn7OUmZkJvV4v/7Rp08a5F0FERORrMjIAvR4oKxN/V7kn1xePCXyWLFmCL774Al9++SW2bduGhQsX4s0338TChQtrfc5JkybBYDDIPydOnHBgj4mIiHzYhg1iWmvWrCttQ4aI7S7k2CSbEz377LPyqA8AJCYm4vjx48jMzER6ejqioqIAAEVFRYiOjpZfV1RUhG7dutk8Z3BwMIKDg53edyIiIp9x7hzQqpX4aO7vv8V2F/OYEZ+ysjL4+Sm76+/vD5PJBACIjY1FVFQUsrOz5edLS0uxZcsWJCcn12tfiYiIfNIzzwBhYcqg56uvAEFwi6AH8KARn6FDh2LGjBmIiYlBly5dsH37drz99tsYPXo0AECn02HChAl45ZVXcNVVVyE2NhZTpkxBy5YtMWzYMNd2noiIyJtt3Aj066dsS0sDfvwR8HOvMRaPCXzmzJmDKVOmYMyYMSguLkbLli3x6KOPYurUqfIxzz33HC5cuIBHHnkEZ8+eRb9+/bBy5Uo0aNDAhT0nIiLyUufPA61bAwaDsj0vD3DTCUMes45PfeA6PkRERHZ6/nlg5kxl26JFwOXZ1vVJy/3bY0Z8iIiIyA1s3gz06aNsu/FGYOVKt0tr2cLAh4iIiGp2/jwQEwOcOaNsP35cbPcQ7h+aERERkWtNmgQ0bqwMej77TJyt5UFBD8ARHyIiIlKzZQuQlKRsGzQI+Plnj0hr2cLAh4iIiJQuXADatgVOn1a2//WX2O7BPDNcIyIiIueYPBlo1EgZ9Hz6qZjW8vCgB+CIDxEREQHA778DvXsr2wYOBFavBvz9XdMnJ2DgQ0RE5MsuXADi4oDiYmX7sWNAu3Yu6ZIzMdVFRETkq6ZMEdNa5kHPJ5+IaS0vDHoAjvgQERH5nj/+AK67Ttl2/fVAdrZXpbVsYeBDRETkK8rKgPbtgYICZfuRI2K6ywcw1UVEROQLXnoJaNhQGfR89JGY1vKRoAfgiA8REZF327oV6NVL2davH7B2rdentWxh4ENEROSNysuBDh2Av/9Wth8+DMTHu6ZPboCpLiIiIm8zfToQGqoMehYsENNaPhz0ABzxISIi8h7btwM9eijbkpOB9euBAN7yAQY+REREnu/iRSAhATh+XNl+6JA4i4tkTHURERF5shkzgJAQZdAzf76Y1mLQY4UjPkRERJ5oxw6ge3dlW+/ewIYNTGtVg58MERGRJ7l4EejcWdxLy9zBg8BVV7mmTx6EqS4iIiJPkZkpprXMg55588S0FoMeu3DEh4iIyN3t3Al066Zsu/ZaYNMmprU04qdFRETkrioqgC5dxL20zB04IC5OSJox1UVEROSOXn8daNBAGfS8956Y1mLQU2sc8SEiInInubnA1Vcr23r2BDZvBgIDXdMnL8LAh4iIyB1UVACJieKig+b27wc6dnRNn7wQU11ERESu9sYbYlrLPOiZPVtMazHocSiO+BAREbnK7t3iKI+5a64B/viDaS0nYeBDRERU3yoqxOnp+/cr2/fuBTp1ckmXfAVTXURERPXprbfEtJZ50DNrlpjWYtDjdBzxISIiqg979gBduyrbEhOBP/8EgoJc0ycfxMCHiIjImSorgR49xMDH3J494p5bVK+Y6iIiInKWWbOA4GBl0PPWW2Jai0GPS3DEh4iIyNH27hW3mjDXuTOwfTvTWi7GER8iIiJHqawUV122DHpyc8VRHwY9LsfAh4iIyBFmzxbTWrm5V9reeENMa1kWNZPLMNVFRERUF/v2WdfrdOwI7NwpBkLkVjjiQ0REVBtVVeIihJZBz65d4ho9DHrcEgMfIiIird57T6zX2bnzStvrr4tpLcstKMitMNVFRERkrwMHgIQEZdtVV4l1PRzh8Qgc8SEiIqpJVRXQq5d10LNzJ3DwIIMeD8LAh4iIqDrz5olpra1br7RlZopprauvdl2/qFaY6iIiIrLl4EFxdpa5+Hhg925xk1HySBzxISIiMnfpEnDdddZBz/btwOHDDHo8HAMfIiIiyQcfAIGBwB9/XGl75RUxrdWtm8u6RY7DVBcREdHhw+LsLHOxseKeWxzh8Soc8SEiIt916RKQlGQd9GzbBhw9yqDHCzHwISIi37RggZjW2rLlStv06WJaq3t31/WLnIqpLiIi8i1HjgDt2yvbYmLEbSZCQlzTJ6o3HPEhIiLfcOkS0LevddCzdStw/DiDHh/BwIeIiLzfRx+Jaa1Nm660vfSSmNbq0cN1/aJ6x1QXERF5r6NHxUUHzbVqBRw6xBEeH8URHyIi8j5GIzBggHXQ88cfwN9/M+jxYQx8iIjIu3zyCRAQAPz225W2qVPFtFavXq7rF7kFprqIiMg7HDsGxMUp26KixFlcoaGu6ZOHycotwNy1RzAmJR5pidGu7o5TcMSHiIg8m9EIpKRYBz2//w4UFDg06MnKLcDQORuQlVvgsHO6k7lrjyA334C5a4+4uitOw8CHiIg816efimmtdeuutE2eLKa1rr3W4W/n7YHBmJR4JLbSY0xKfM0HeyimuoiIyPP89Ze4l5a55s3FdFfDhk572zEp8XIqyBulJUZ7bYpL4lEjPvn5+Rg5ciSaNm2KkJAQJCYm4s8//5SfFwQBU6dORXR0NEJCQpCamopDhw65sMdERORQRiNwww3WQU9ODlBc7NSgBxADgx+e6Of1wYE385jA58yZM+jbty8CAwORlZWFvXv34q233kJ4eLh8zMyZMzF79mzMnz8fW7ZsQcOGDTF48GBcvHjRhT0nIiKH+OwzMa21Zs2VtkmTxLRW796u6xd5FJ0gCIKrO2GPiRMnYuPGjfjNfHqiGUEQ0LJlSzz99NN45plnAAAGgwGRkZH49NNPce+999b4HqWlpdDr9TAYDAgLC3No/4mIqJaOHwfatVO2NW0qprsaNXJFj8jNaLl/e8yIz/Lly9GrVy/cddddaNGiBbp3744PP/xQfv7YsWMoLCxEamqq3KbX69G7d29s3rzZ5jkrKipQWlqq+CEiIjdhMgGpqdZBz6ZNwKlTDHqoVjwm8Dl69CjmzZuHq666CqtWrcLjjz+OJ598EgsXLgQAFBYWAgAiIyMVr4uMjJSfs5SZmQm9Xi//tGnTxrkXQURE9lm0CPD3B7Kzr7Q9/7yY1kpOdl2/yON5zKwuk8mEXr164dVXXwUAdO/eHbt378b8+fORnp5eq3NOmjQJGRkZ8u+lpaUMfoiIXOnECSAmRtkWHg7k5XGEhxzCY0Z8oqOj0blzZ0Vbp06dkJeXBwCIiooCABQVFSmOKSoqkp+zFBwcjLCwMMUPERG5gMkEDB5sHfRs3AiUlGgKerx9kUGqG48JfPr27YsDBw4o2g4ePIi2bdsCAGJjYxEVFYVss2HR0tJSbNmyBckcFiUicl9ffimmtX7++UrbM8+Iaa0+fTSfztsXGaS68ZhU11NPPYU+ffrg1Vdfxd13343ff/8dCxYswIIFCwAAOp0OEyZMwCuvvIKrrroKsbGxmDJlClq2bIlhw4a5tvNERGTt778By/ICvV5MdzVuXOvTevsig1Q3HjOdHQB+/PFHTJo0CYcOHUJsbCwyMjLw8MMPy88LgoBp06ZhwYIFOHv2LPr164e5c+eiQ4cOdp2f09mJiOqByQTceiuQlaVs/+03oF8/1/SJPJqW+7dHBT7OxsCHiMjJvv4asFxXLSMDeOst1/SHvIKW+7fHpLqIiMiD5ecDrVsrmsqCQ7BxzQ7cmGzfqDyRI3hMcTMREXkgkwkYOtQq6Hl+/HvoPOEbzP6z2EUdI1/FwIeIiJxjyRJxttaPP15pmzABEASkPHgHElvpWYBM9Y6pLiIicqyTJ4FWrZRtISFAYSFwuf4iLTGaO5x7qazcAnlWnTt+xxzxISIixxAEYNgw66Bn3TqgrEwOenydMxdYdIfFG9XWUXKHvgEMfIiIvFK932S++w7w8wOWLbvS9uSTYjA0YED99MGBnPn5OXqBRfO+OnPxRns/kzEp8TbTmO6ysCQDHyIiL1RvN5mCAkCnA+6880pbcDBw9izw7rvOfW8VjghanPn5qQUGtWXeVy3n1vo52fpMZqzYi05TVmLGir1yW1piNH54op9VmsvR111bDHyIiLxQUlwEQgL9kRQX4Zw3EATgjjuAli2V7WvWABcviisw14LazVjLTdreVEt153SXm7Q9zPuqFnTYojW4s/WZLMrJQ3mVEYty8mp8fVpiNJLiIpCxZKciUKpvDHyIiLxQztESlFcZkXO0xPEn/9//xLTW0qVy0/F7H8DQ2b8hq2nHOp06M2s/cvMNyMzab1e7LfamWqTfM5bstAp+tAQQgLbATMu12ENrXyVag2Nb7zMyKQYhgf4YmRRTzSuv0BIoOQsDHyIiL+SUEYvCQjGtNXz4lbbAQODMGYzr86Bb1G9Ux/IzGZMSj5BAf5RXGevc79qkxsoqL7m02NcRwfHkIZ2xb/rNmDyks13Haw2UnIGBDxGRF6rtKIBNgiDW8ERbnCs7G6isBJo0qTHQsndEZFJaAhJb6TEpLcGudlvURlQsP5O0xGi8ffc1Vv2uTY2QlkBTupbQoAA5WNL6no6oY3JFOk9roOQMXMeHiHyOu68z4la+/x74178UTT/1vR26efMUn530Z2nEw/JzNR8Rqe4zr+/1fWy9n719rek8NRncJRI5R0vk3eS1vGdt+uiIPnsDjvgQkc9xl2m1bq2oSExrmQc9fn6497WfMKbfwzY/O1ufa1ZuAQbMXIMCQzliIkJrNbqQlVuA8Yu3y6M4NY12qI0OaZmOHRMRCkN5lVOns+ccLZFHoLSOvrhitMZd1uGpKwY+RORzPGnGTr0TBODuu4GoKGX76tWA0Yj0W7qpfna2Pte5a48gr6QMp85XQh8SWKsRhrlrj6DSKAAQ62JqClzV0nz2BrxpidHQhwQir6Ss3qaza01NOjSVaSdv+QcDAx8i8jmuuGk4k73/Eq/xuGXLxNla33wjN315zc3o9GIWkJoKoPrPztZz0uhJbUd7pHP46cQ/V1wy1Tpw1fI6rTOetIyGqH2GWs7hKzVBzsAaHyLyar5Qz2NvvYfqccXFQGSk1fFvfbURH+0pxcikGMXnKJ3Lns/UEXUkaYnRCPDTodIooKLKWOtzanmdlhlPWbkFyFiyU54dVtvr1VK3w5qg2uOIDxF5NW8Znq+Ovf8StzpOEID77rMOelatAgQBT9/bR56BY/45umIvpvQ+7RAS6I/0Pu0cfm5btIxuzF17BOVVRoQE+tdpNMTWe6p9pp5QE+SudIIgCK7uhLsoLS2FXq+HwWBAGDfTI/IKvjDiUys//ADcdpuy7cEHgQ8/FIuaLUifY1JcBJZuz0dp+SWk92mrmJY8dM4G5OYb5Busqz53W9+5o/4eOPPctph/pj880c+h5/YmWu7fHPEhIq/mTfU8av/6t7Vfkqp//hEDG8ug5/Rp4KOPbAY9wJXPMedoCU6dr0Sl0YRVe4oUx5iPKtQ00uaI0SG1c9hax8dRI3+2zuPMv2McqXE8Bj5ERB5CbWE+u7YBEARg5EigRQtl+8qV4nMR9hXxjkmJR5C/7eDI8rjqbtiOCES0nKO6/mgJwpwViKj1wZsCd3fBwIeIyINl5RYgwF8HHYC+7ZvaPmjFCnG21hdfXGnqdTNgMgGDB2t6v7TEaAzo0Bw6AB0iGymeMw/MarphO2IXcbVz2FrHp7r+aAmgnBWI+EItmrvgrC4iIg8xKS1BMbMKEG+Y5y5eAgAUlVYoX3DqFNC8udV5uj35JQKaN8MQlbRWTTYePg3h8mNtaZkhpDYjSe0cttpnrNiLRTl5GJkUo6hLysotgKG8qtrp9o6u4bF1PvOaKHIujvgQEbkZLWmPMSnxaNxA/DdsZFiw2CgIwL//bR30/PQTBrz+K86GhCE0SPnvXi3pHrWNJrXsp6WF1vSSrWtRSwdKCyxWt7iio0dj6rtOiJQY+BCRV/PEZfbVannUnL884rPx8GkgK0tMa33++ZUD0tPFtFZamiI4Mf9stNzc1Taa1HLzVvtebLWnJUYjKS4CGUt22lXAbeta1II1taDKvB+OruthwbJrcTq7GU5nJ/I+njgdeMDMNcgrKUNMRCjWPzew2mOl64soM2DbnBFWz/+yLhepA7pW+1pnTz+3ldpR+17U2jtNWSmvlbNv+s2a308rT/x748s4nZ2I6DKtWw/UhaNGl+xNGWXlFsBQVonZP8+2Cnr+++hMtHv+R7y786zq6+tr5MHWCIzae9v6vrJyC9Aw2B9B/jqrERtbtGwJobVw2hNHEEmJgQ8ReTUtWw/YouVGV5taELXUjj0po83zvsT6iYNw2/afrzSOGgWYTOg9dmSNQY35+1j23bxfdb3ZWwYRWbkFyMzaD0N5ldWxq/YUobzKqFgjaO7aIzh1vhL+fn7oERNeqz5I57H8fqS2jCU77foOOPvK8zHwISKvVtdRDS3bM9TmvWp1Iz19GtDp8PK8ZxTN2Wtzgc8+A3Q6zcWyln23Z4sKe1n2RSootnf38zEp8QgJ9Jf3wqotWwGYobwKQf5+dp+b9Tmej9PZicir1XVjRbVpxramWNfmvbRMY87adRL+jz6Cm3JWKNpffuQ1XDvu33W6Tsu+W/bL/M91raEZkxKPKct2o7S8yioFaWvKPgA0bxwsv7a2LK9RCsBiIkKhDwm069zeslGnL+OIDxG5JbV6jAEz12DAzDX1VmOhNnLiiH/5z1ixFxlLdiIpLkJxfpuppdWrkXZNK2XQc//9gMmEqR88b1c9S3Xtlsyv29aIjZZNSi3b0xKjEa0PQaVRsCsFac+U89qQvsNJaQlW3zFrebwXAx8icktq9RhaUiTOZG8qqbobaHVry8jXXlIi7p91003KFxcViSsxqyxCaOvzy8otQMaSnXWuUVEL+tQCIi3FzWrHxkSEwlBe5dBApDarOWspkCb3xMCHiNySrRujdAO0tcquu9x8LPtRXX1MTWvLvL9uPtDUYhuKZcvEBQov77mlNgpm6/Obu/aIPCW8rikjWwGD+YysmtbB0TKSlpYYDX1IYL0GvFoCMxY8exau42OG6/iQL3H0MvyuVt/rrqh9fpZr8EjHJcVFIOdoiX2f9y+/ADfeqGy75x7gq6+sRnik6wZQ47VX95074u+D+bXrQwKRm29ASKA/3r77mjr/Hatr/xz1993WebztvyVPxHV8iKhG3vavVFv/QlcbBZqxYi86TVlp1yrAatQ+v7LKS4pHaWQj52hJzZ/3mTPiqsuWQU9hIbB4MWb8tM+q39WNglmqbgVkratF18RRM7Ekdd3SwVF/3231g9tNeBYGPkQ+ytum5dq6+ajd7NRqa7RQ+/wqLpkUjzUdL3vsMSAiQkxjXbb1nY/F3yMjAQALN/2F8iojFm76Sz4mLTEa658biPXPDbSrONfWtc9YsRcnSsrsv3gV5gsvpiVG4+27r3Gbv2Pe9vedao/T2Yl8lC9My1WbKt63fVNk7ytG3/ZNVV5ZO1m5BfJO6eWVRsVzqp/3r78CgwYpmjZ0G4iRN2Ug0a8JfjBrDwsJxKnzlQgLCayxL2o7mo9MipF3KZcsysmDAEAHOHSDUXf6O+ZOfSHX4ogPEXkttRTEwaLzEC4/1taUZbuRm2/AlGW75TbzNFFIkH/1Jzh7FqaAAKugBwUFOPfZF0hs3cQqYPtX91YICfTHv7q3UrRrWUzR1gajUpH1Q/1j6xQceFv6lLwTAx8i8mq1ne1VUx1QafklxaOl8NAg9W0fxo4FwsPhZzQbFfruOzGtFRWlGrDZ2s4BsB1waKk7UdttXSumk8gTMPAhojpx12nkEltBgT2bgNZUB5Tepy1CAv2R3qet4ryNGwRAB6BDZCOrbR8abfoNaVe3BObOlV+z6ZrrkbUzH7jjjlpeed0DDkd9hyzyJU/A6exmOJ2dSLv6nkautR+1nWo8Y8VeuRZGy0hIpykr5bVy3r77GsxdewTXRwZi/P39EGi0GB06eRKItr9Pzpo27S7fIVFtcTo7EdUbLaMNzhwdUuuHllEI8/7VNv1jvihhWmI0fji0BM/ck6QMer79Flm7TmLot0dsfhZqn1NdR1TUzmu+8CCRt2PgQ0R1ouVm7KziV0eNhDhiLRs5YGr0j7jY4Jw58nOFN9wMmEzA8OHVfhZaPictwaTa9eUcLUF5ldGufbOIPB0DHyKqN84aWZAChYwlO60LiS04atRJ9TwGA9CgAZCSomzPz0dUdpa88nJ1I2VaPidHBJMsSiZfwsCHiOqNs0YWLFcJtmc0RQqSzNlT9CyxNZ0dEyYATZoAFRVX2r7+Wpyt1bKl4vVpidHyOkOW/VD7nLRMW7dF7fpsjdo5PUAkchEGPkRUb5w1smC5SnB171PdVgpa0naK6ezr14sjOe++e+WAoUPFtNbddwOwHQCoBWhqIz62UlXVBVCWXJGW5No+7stXg1IGPkRUa1r/x1lf053N3+ehhX8gduIKPLTwD/k5R2ylkN6nLZqZLmL3W8OB669XPHfHC18Dy5crNhS1FQCoBWhq6/VY7gNW3bnrylFBKtNo7stXg1LNW1ZcvHgRc+bMwZo1a1BcXAyTSbkfzbZt2xzWOSJyb2rbIrhTX7L3FUO4/ChxxPYFk7M/wuRZsxRt/71/Cj5t0xsx+lCr421tn6G1H6FBAQAqLz9anzspLgJD52xwyHR3R23xwK0i3Jfali7eTnPg8+CDD+Lnn3/GnXfeieuuuw46s3/REJFvcaf/car1ZVCnFsjeV4xBnVo45o3y8sRanqVLr7QNGQIsX47ee4qw1QGfR4fIRjhRUoYOkY0U7ZPSEmxeoxRcSOvxuEMgSu7PV4NSzQsY6vV6/PTTT+jbt6+z+uQyXMCQyH3UdgFBR5Omyo/r0xqDf/4KmD4dKC+Xn3/gpe/w6dSaV13Wskig+SKI+6bfrLmvjl7gkOoHv7/ac+oChq1atULjxo1r3TkiInuobRlR0x5adaFWgNxkwxp0vKkf8MILYtDTvz9++zYbQ2f/hnv+lWzXuW3VumTlFmDAzDUYMHON/J5ZuQVoGOyPIH+dYgd1e/oNgFtGeDBfrbmpb5oDn7feegvPP/88jh8/7oz+EJEPqa442nwFZPPjatpDqy6sZk3l5eHD5a/h8yVT0a4kH6cbhQOffw6sW4f+w2+wGWRoKfieu/YI8krKkFdSJt/s5q49glPnK9ExKszukS7eML0DC8Hrh+bAp1evXrh48SLi4uLQuHFjREREKH6IiOxV3Q27R0w42rdohB4x4YrjzAMiZwm4VAW89hrQqROiflmBSzo/fNzrdox8ZiEwciSg02naFFWtfUxKPGIiQhETESrf7LTe/LJyC2Aor1Kcw9356jTqmnCT1/qhubj5vvvuQ35+Pl599VVERkayuJmI7GZZw1BdcbR5oJAUF4HDxeeRFBeByUM6a6r5sbduQroJ33FqL17++gPgLzFA2RN3NZ6+/mEcb9Ueb99xjc3+mZ9X7ZrMr0Fiq7hUa8GpNGqU2ErvMTdMd5oNSL5Hc+CzadMmbN68Gddcc03NBxMRmcnM2o+8kjJkZu2Xb/BpidHyCIB5cGIeQMxde6TGFZ/VAhx7b7JLlm7C81/OwpADG8WGyEjgjTeQd80glK08gOYWx5v3z/K9bb2PM1etdpeZdfbyxD6T99Cc6kpISEC52YwGIqK6spUGMh/2l9I/0jo1tlIkaimmGlNHlZXA66/jw8xRGHJgIwQ/P2D8eODAAWDUKKRd3RL6kEBFHY5l/+ypsXHmqtWelh7R2memxsiRNAc+r732Gp5++mmsXbsWp0+fRmlpqeKHiEiN2l5RNQUF0o0y52iJIsAwvyHWKrDIzgauuQaYOBEBF8uBvn2h27YNeOcdQK+vtn9a39vem73aTd6Xb/71Xbzty5+1L9C8jo+fnxgrWdb2CIIAnU4Ho9HouN7VM67jQ+Q+bK3jk5VbIM+4khbzq2ltHFvr56xZvRUBzz+H/tt/FQ9q0QJ44w1kdUvF3HVH7VpHRcu6PFqonbe69/P29V/q+/qc9d2S8zh1HZ81a9ZgzZo1+PXXXxU/Ult9ee2116DT6TBhwgS57eLFixg7diyaNm2KRo0aYfjw4SgqKlI/CRE5hSP+xWxr2npaYrQi7aSWAjN/f8WGn5WVwBtvoPet/dB/+68w6vyAJ54Q01r//jfmrjtq98iC2kaidaU2elTdqJK3T2ev73Qep5V7N83FzddbbMbnCn/88Qc++OADXH311Yr2p556CitWrMA333wDvV6PcePG4Y477sDGjRtd1FMi76HlX92OmLUzMilGHvEx74P51G2pkHjAzDXIKynD+MXbrd7fUF6F8ioj/lmWBTz7MbBvH0IB/NmqE96+/Ul8Ofsx+fxaim6dVaysVhxd3Wwvdy8W9rQRKV/dysFX2JXq2rVrF7p27Qo/Pz/s2rWr2mMtgxFHO3/+PHr06IG5c+filVdeQbdu3fDOO+/AYDCgefPm+PLLL3HnnXcCAPbv349OnTph8+bNSEpKqvHcTHURqdMy/K/1Rmfv8Wp9kAIfAPK/1KXzzf58Lcb8MA9D9/8mHty8OV4dOBoftuuLNk0bYf1zA2u+eBt9NP8dgDztPudoicfc4OsLU0fkbA5PdXXr1g2nTp2S/9y9e3d069bN6qd79+51730Nxo4diyFDhiA1NVXRvnXrVlRVVSnaExISEBMTg82bN9s8V0VFBYuziSyopam0DP/bk5owfx9bqRpb/VDrw+AukQjy90OzRkFy0PHDY72RtuoLfPvugxi6/zcxrTVuHHDwILq/OB5dW4dbFVlXx7KPtmZ1LcrJ8+qUU21Zfm8sHiZXsivVdezYMTRv3lz+s6ssXrwY27Ztwx9//GH1XGFhIYKCgtCkSRNFe2RkJAoLC22eLzMzEy+99JIzukrksdTSVI4e/jd/n8iwYOzOByLDglX7Ud2Iy6o9Rag0mlBaXiW+eM0aMcjZuxcNAWxtmYD378rA/73zsHgtTZpovpbq0knSc+YjPnSF5d8dLmBIrmTXiE/btm2h0+lQVVWFl156CSaTCW3btrX54ywnTpzA+PHj8cUXX6BBgwYOOeekSZNgMBjknxMnTjjkvESewBEjO1rPrfY+Gw+fhgBg4+HT8ustt2GwHHGRfs9YshNllZcAAE3OnkLAv0cCN9wA7N2LU6F6PHPLBNw1cibuevDWOl1Pdak4afRn8pDOHremjiuweJhcSdOsrsDAQHz33XfO6ku1tm7diuLiYvTo0QMBAQEICAjAunXrMHv2bAQEBCAyMhKVlZU4e/as4nVFRUWIioqyec7g4GCEhYUpfoh8QVZuATKW7LSZlnHEDBrzIEUtCDJ/n77tm0IHoG/7pvLr80rKoA8JlPshzaKKDAvG0Dkb5N/Lq4xo7A889udS/PrRY7hxx6+Anx9+7H8Hbnj4A3yXmIqIxnX7x5K7z5rytNSRJy66SN5D83T2YcOG4fvvv3dCV6o3aNAg5ObmYseOHfJPr169MGLECPnPgYGByM7Oll9z4MAB5OXlITk5ud77S+TOpC0gQgL9nfKvbvN/0asFDeY364NF5yEAOFh03ur1EmkW1cbDp5Gbb0DO0RK8ffc1uK/sKJZ/PA4Tsz9Go8pynLm6J/DHH/B//z20jW+FNhGhOHW+sk5Bi62p6/UdbFT3fu4emBG5E83T2a+66iq8/PLL2LhxI3r27ImGDRsqnn/yyScd1jlzjRs3RteuXRVtDRs2RNOmTeX2Bx98EBkZGYiIiEBYWBieeOIJJCcn2zWji8iXmNerOONf3bZqOiz3tJL27Xruu104d/FSta83T311iGyEjYdPI1V/CWmvPYO0L78EABga6pH33FQkvjgB8PND2uXXZmbtr/PO5bamrqvVqThr6nZ1dTHuPp2dyJ1oXrk5NjZW/WQ6HY4ePVrnTtkrJSVFns4OiAsYPv300/jqq69QUVGBwYMHY+7cuaqpLkuczk7kGGo3f/NpzYbyKnkKuiS1UwsUlVZU+zp/4yX0XL4IT2/6CqEVZYBOhxV9h+GFHncjpn1rxXTpXq+sxqnzlWjWKAh/vnijQ6/Hnmt05NRtT1sLh6g+abl/aw58vBkDHyL7WW4fAcBqNCcmIlSxTo75zXtb3hksyslD88ZByCsRNz6WanYsgwbpdS82KkbnGS+g8ZED4hO9ewPvv4+soJY2g4L2L6zAJRMQ4AccfnWIsz8Sq2vUGqAwuCGqHaduWfHyyy+jrKzMqr28vBwvv/yy1tMRkYvUtUZFKkCWto8wT8WcKasEAJwpq1S8j3lRq5Q++ueceGxIoD9GJsXYnO2T1gz4Yct89B49XAx6mjYFPvoI2LQJ6NlTtVi2SWiQ4tHdsVaHyPk0j/j4+/ujoKAALVq0ULSfPn0aLVq04CalRB5Ca0rG1jo6aiM+Y7/cBpMA+OmALi31yM03ICYiVD6X+fFJcRFYtadIbjdPJc3PPoDXTq5Hp/lvAufOATod8NhjwCuvABE175Fl2Ud33+CSIz5EtaPl/q25uFnahd3Szp07EWHH/4iIyD2oFcSq1bNkLNmJ8iqjXFxra0FD6Xfpn1OCcOV9zGt6MrP2Y/1zA+Xjl27Px6nzlZiybLfctu6j7/D6l28i4dRxAMDZrt3x0s1jcNO/hyBNw/9r/jlXoei3s9Wl0Jh7RBE5n92BT3h4OHQ6HXQ6HTp06KAIfoxGI86fP4/HHnusmjMQkaM4c2RAqs/JzNovn1ua/q4D7NqN3M9PB6NJgJ+fTr6ZP7TwDznwkRYclK7l1Hkx3XW2rBIoLASefRavLVoEADjXSI/Gb7+Bf5d1xK6CczisIYBx9rR9W5wZvHBEiKju7A583nnnHQiCgNGjR+Oll16CXq+XnwsKCkK7du24Xg5RPXHEkv9azjEmJR7jF29HpVHAqj1FmDykc7XHD+zYHNn7ijGwY3O5TVqV2VY/AMDfZMRjuSuBjvcDpaUwQYevug3Gezf8B5sfvhuPW2wKag9nT9uvK62BDLd6IKo7uwOf9PR0AOJ09r59+yIgQHOWjIgcxBHrtqidY3CXSCzKycPgLpFyW1pitDwSZE7txm25ICEAjEyKwYe/We/1NyYlHivmLsFT389GfOHl5TB69cKrt4zDxxXNMKhTC7kPUm3R0Dkb3DaY0UJrIMP1eojqTvOsruuvv55BD5GLqc1iUpupZatd7Ry2FusDxOLgxFZ6xY7marOQzGd1SSYP6SwXOIcGXf5/SGEh0t54Hu/NnyAGPRERwAcfADk52NIsHgKAotIKxbm1zHxy91lSWves4lYPRHWnOfAhIufQErSoyczaj9x8gzyTSWIrAHDEJqXSbup/nb6gOE9FlVHxKBncJRIhgf64OaEpMHs20LEj8PnnEHQ6rOxzG375fj3wyCOAv7/NbSK09s/dN8NkIENU/zh0Q+Qm1NIeta3rkNJQSXERVjudA7aLmLX2T6rbOXfxkqI9St8AeSXliNIrNwfNOVqCLsd24b7/GwecvByE9eqFx5L+g1UN2yJmczFS+3eRj7U18qSleJizpIjIEkd8yOu4+07VWkdatIxamKejpEBlUU4e8krK8M85ZcrIVjoKUE8P2eqHtJu6n0452+vE5ZWYpUcAQFERXln2Jr794nnEnjwChIcD8+cDOTnY1DTeqi+OGq1x978PRFS/GPiQ13H3ug61/qmlPWy1q93Mt+WdweHi89iWd0YOHPq2bwodIK9lI7lQcUnxKFELONISo+XiWul9pfobkwDFyEybiJArj5cuAXPmAB074prsZQCAVcm3AgcPAo8+iqy9xTh/eZNS89SYo9JA7v73gYjql+bA58KFC5gyZQr69OmD9u3bIy4uTvFD5GruXtfhiP5Z3sylQOjjDcdQXmXEJxv/kgMHaYYVoL4GT02jItLzUg2R9L5jUuIRExFqlUaTtqFovXc70KsX8OSTgMGA/a064NGx78P0wQKgWTP5WgQAOgDpfdrV+jNxRM0SEXk/zTU+Dz30ENatW4dRo0YhOjra5irORLXliAXa3L2uwxH9s5zWLAVCEinQycotQKHhSrrJfA2ehsEBOHfxEhoGB1gFUpa1PNLzQf46RZCjdi09Qypx+/dzcNfuX8SG8HC8f+NovNV2AIKDgzBM5Vrq8rmo1UK5+98HIqpfmgOfrKwsrFixAn379nVGf8jHcYE2+1jezKXgITIsGBsPn8bIpBgA4udZaRSgw5VgSBIc4Idzlx/Ngw8pXWY+OmS+gKF5IGXl0iVg/nzMfXkiwiouiG0PPghkZiKu8BKCLba9sHUttcU1bojIHpoDn/DwcO7JRU7jCzcvtVEtLaNdM1bsxaKcPIxMisHkIZ1Vgwfp80yKi0DO0RL5c83KLRC3h7jM/PXSNg/mNTvmCxhWGgXbgenmzcCYMcCOHQgDsDsyHssffgEvTB8tnuPyIs7O+n45skNE9tBc4zN9+nRMnToVZWVlNR9MpJEvrGuiVmyrpQh3UU4eyquMWJSTp2i3rHORCpJX7SmCobxKPiZjyU5cMomvkRcTvExt/ZxJaQk263my1+3Gz8m3An36ADt2AE2a4K1h43Hbv9/GysaxinP4wvdLRO7NrhGf7t27K2p5Dh8+jMjISLRr1w6BgYGKY7dt2+bYHpLb48aJ6mx9NmqjWlpGu/q2b4rsfcXydHKJ2gaj5rui60MCUV5lRJC/H6L0DeSVmKW+FhjKUV5ltNqTy2pExWgEPvgA1z3zPBqXX96aYvRo4LXXsOyT3TCV8B9HROR+7Ap8hg0b5uRukCfz1LocRwZsauey9dmopWTU2s3PLZ2zwFAu74Vl/rzl2jxZuQUwlFfJNT5nyirlNX6keqBteWeQlhitKGCuUU4OMHYssG0bGgPY0yIO7935FOa9Pw4A5Pfw5pQlEXkmnSAIljWPPqu0tBR6vR4GgwFhYWGu7o7H8NQRn6FzNiA334DEVnr88EQ/p5zLEZ+N+bkByMFJpVFATEQo9CGByM03ICTQH1VGIy6ZgCB/HQ7OuEV+rcRPBxzNHAIA6DA5C5VGE4L8/XBwRppipWepHsiyBmnRD3/itT+/QpulX4mNTZpgz5hn8UKLvnjshg41Bm6e9PeDiDyHlvu35hqfuLg4nD592qr97NmzXMfHR9VH3YYzVt+tbn0Xre9X3aJ/9i4+qNZuXnMjvc+ADs0REuiPwV0iMSYlHiGB/iivMiIkKAA6AAM6NFf0y+/yII55yjosJEDxKOkRE279fRqN2Dv1dbw//f4rQc9//oPspesxMWqAatADcAFBInIvmgOfv/76C0aj0aq9oqICf//9t0M6RWTJGTfP6gI2re+nJfjTWty8dHs+yquMWLo9X7EooVSHk5YYjbfvvkYeERIA7DhxVtGvGxJaQAdgYEcxIJKCqyB/P/yreyvF+2cs2akMvrZsAa67Dk9//y6aXDyPA1HxwMaNwP/9H97JNdT4OdV2AUFuNUFEzmB34LN8+XIsX74cALBq1Sr59+XLl2Pp0qWYPn06YmNjazgLUe1IKwQbyqvq5UbozNV+te7JVXp5Npb0aIsU4FRUmS4fq9yGoqi0AgKubDExd+0RnDpfiUqjCUu358vvL40czV17BDh1Cnj4YSApCdi2DecaNMTU1Efx2JPzxRlc1fTZPGip7YggR4qIyBnsXsdHKnDW6XRIT09XPBcYGIh27drhrbfecmjniCTmxbe1LaLWUmviqDVhHFHfkt6nHRbl5KFv+6YYOmcDxqTEWxUPZ+UWIDNrP4ID/S6/pq3iHJYzxsakxGPMF9sg4EpAJfVv/q8HMeOfTUCH24AzZ8QTPPAAfk9/CttzS/GcWZCj9jk5ouDdF9Z0IqL6Z/eIj8lkgslkQkxMDIqLi+XfTSYTKioqcODAAdx6663O7Cv5uLqOwqiNIDgzpWLrPbWmuiYP6Yx9029GUWmFIpgwH0WRpqyfu3gJVUZTjf1KS4zGoE4tFPVAAJBWfgILPxyPrtOfF4Oea64BNmwAPvkEg1KutnvkxhEjZlzzh4icQXONz7Fjx9Ds8uaCRPWprjdCtZuxrYDDUcGQrcUAbfVDmnZuuTigWv9nrNiLTlNWYsaKvfJzAZf/axYALNz0l+K1U5btRm6+AVOW7Zbbdpw4e6Ue6NQp4JFHgKQkNNmzE6VBofhg+Hjgzz+BWmxPw6CFiNyVXamu2bNn45FHHkGDBg0we/bsao998sknHdIxIkeraVsH84BDa6pGLaWVc7TE5vYP0jHS6wzlVcgrKUNiK73q+5m/Tto3a+Gmv+QtK5osC8Kp8+L6PWEhyoVFbdUJlZZXwc9kRNqm5cCse4ESsY9/D70Lk3qPxP23XQsEaN7VhojIrdn1f7VZs2ZhxIgRaNCgAWbNmqV6nE6nY+BDHsdWQKS1vkQtULI8j1SLA4iL/EkrLTdrFCQXb89YsVexr5b58dK5/fx0gFEQHy8Tt56oRJC/DtNv76ron1QnJG1eCgATm51Drzen4uqTB8WGq68G3n8frfv1w+d2XbU6rt1DRO6KCxia4QKGzuetN0R7r8t8QcHEVnp5pMdyIcLyKiNCAv3RvHGwvN2E+eKIHSb/hEqjIC9UWF0frNpPnwZeeAH48ENAEICwMGD6dHGD0YAAh2yi6sjFIYmIauLUBQyPHj1a644R1ccUZWcWK9f13NK0fKmWZ1JaAhJb6TEpLUGu4RmZFCMHPwBsbgya3qcdQgL9kd6nndwmbUg6d+0RRf+kz/zpr7cjd9qbQIcOwIIFYtAzahRw4ADw5JNyWssRm6g6czkAIqK60Bz4tG/fHjExMRg1ahQ+/vhjHD582Bn9Ii/liBtiTcGH1uBKSzCTmbUfufkGOf1k73tKxcjb8s5g/XMDsf65gXKKTSoC/m7b39idb8CxUxfkBQkHd4mEPiRQkeYCxNWV27dohB4x4TX2b0xKPHoUH8aX//cUEl9+VqzlSUwE1q0DPvsMiIpSnEPrOkO2sLiZiNyV5sDnxIkTyMzMREhICGbOnIkOHTqgdevWGDFiBD766CNn9JG8iCNuiDUFGVqDK0eMQtX0noty8lBeZcSinDzVc/yyrxjC5Ufpc1q1p6hWgRYgBnT3vfYTukyfiG8/fQrdCg7iQnAoMGsWsG0bMGCAzdepfUf1Hcxw5WYicgbNgU+rVq0wYsQILFiwAAcOHMCBAweQmpqKJUuW4NFHH3VGH4kUagoytN6gtQRK5qkpy/eU0kwzVuyVb9jSzbtv+6YICfRXFBdb0lk8mis0XFQEAGp9lvs3uAMOv/oO3n/5PsR88xn8BAHLuw7ERx/+BEyYAAQEWE2JdzdcuZmInEFzcXNZWRk2bNiAtWvXYu3atdi+fTsSEhKQkpKClJQU3H777c7qq9OxuNl9eUJRtFTQK9XnmO+mbk+Rb6cpWSivMiEk0A/7pqcBEK9bmroeExGK9c8NrLkjW7cCY8eKe2wBOBffEa/eMhZfhcYp+tFpykq5iHrf9JvrcOXO4QnfORG5By33b82LdDRp0gTh4eEYMWIEJk6ciP79+yM8PLzmFxLVgSO2QHA2acQnKS5CMR29pmnx0g2+yij+G8R84eW0xGh5ynuNSkqAF18E5s8XC5cbNwZeegmNx43DgP2nsNuiHyOTYqymuLsTR20bQkRkTnPgc8stt2DDhg1YvHgxCgsLUVhYiJSUFHTo0MEZ/SMC4Np9m2oaeTB/3nxUx3zNnupIQV2zRkG4UGG0CkQs9+WyYjIBn3wCTJworsAMAPffD7zxBtCyJQDbQcTkIZ0xeUjnGvtHRORNNNf4fP/99zh16hRWrlyJ5ORk/Pzzz+jfv79c+0PkDM4srLVVRGveZj7aZOtY6fmMJTut2vNKypBXUlbtdhhSvc7027ti3/SbtQUj27aJO6U/9BBw6hTOxXfApCdmI2vim3LQQ0REV2gOfCSJiYno27cvkpOTce2116K4uBhff/21I/tGBMD56/KMX7zdauaUebBjXkhsq+B2TEq8XNdj2W6+Bo90HdKUc+lYKagDYPM6bRb5njkj1vH06iXW8jRqBLz1FkaN+wBfhcaxIJiISIXmwOftt9/GbbfdhqZNm6J379746quv0KFDB3z33Xf4559/nNFH8kJaghlnzu6Zu/YIKo3W9f3mwY75aJP5pqPSNWzLO4OGwf4I8veTNyOVRoompSXIa/ZI1wFA02apis1LpbRWhw7A3LliLc9994mLEGZk4NFBHT1y4UBOXSei+qK5xuerr77C9ddfj0ceeQT9+/eHXq93Rr/Iy2kpVnZmfc+YlHjFXlhA9bU55puO5hwtQW6+AYeLz8urLEubkUrX9/gX29CsURCm397VqvjZnuuU0mWJrfRIu1QI9BsObN4MAMiLbIuCGW+i94N3Ou4DcRFPKF4nIu/AvbrMcDp7/XHEVGVnTXc230+rWaMg/PnijfL7TVm2G6XlVUjv0w49YsLlQGbp9ny5ffKQzsjKLcCYL7ZB+o/LfBq5ln2ssnILsPCnHXh9xxK0XfKZOOLTqBH+b9C/8epVN6FTTFPFOTx1jyxOXSeiunDqXl1EjlAfKzjX1piUeHkRwdLyKsX7nTpfiUqjgJyjJfI1TB7SGaFBAag0Cli1pwiAeH0P9Y9FgB/g76eTU2DS+e1KR5lMSNu6CgtfH4W2iz8Vg5577wX270f09BfRKaZpnbaVcCfc4oKI6ovmVBeRuzBPDTlyxEAKWizXuDFPi9kTWEwe0llOh5mntuxan2bHDrF4edMmBAM41LQNPr3vacyYM148RyvYPAfXviEiqh5TXWaY6vI8UsBjKK+Sa2FckeJRC7w0B2RnzwJTpwLvvy+O8DRsiP2PPIVJbW7AI6lXNiplaoiI6AqnrtxM5ApqN3op3RUTEeqyFE91xdB2j8AIAvD558CzzwLFxWLbPfcAb76JhNatsdTicBYDExHVjuYanxtuuAFnz561ai8tLcUNN9zgiD4RWVGr55FqWialJcg1IlqnRtd1KrXaQoV227lT3Ck9PV0MehISgF9+ARYvBlq3tvkST63lISJyNc2Bz9q1a1FZWWnVfvHiRfz2228O6RSRpciwYOguP5qzVRQrLRBoPgqjFtxk5RYgY8nOOhVJWy5UWNN7ygwGYPx4oEcPYMMGoGFD4PXXxUBo0KBq35PFwEREtWN3qmvXrl3yn/fu3YvCwkL5d6PRiJUrV6JVq1aO7R3RZRsPn4Zw+dGcrRRYWeUlxSOgnhqau/YIyquM0AGKmVdqbL2fWjpLNR0lCMCiRWJaq0icBYa77gLeegto08aej4OIiGrJ7hGfbt26oXv37tDpdLjhhhvQrVs3+adnz5545ZVXMHXqVGf2lXyA2ijJyKQYhAT6y7Os1LZ/UCONyhjKqxTnlrabEACbiwpa0jKF3mY6atcuMa3173+LQU/HjsDq1cCSJcg6G8DVi4mInMzuEZ9jx45BEATExcXh999/R/PmzeXngoKC0KJFC/j7+zulk+Q71EZJesSEI+doCXrEhCuOs1XUHBoUAKDy8qPIfMsI83NLj9WtDG0+yqNlFWnFSJDBAEybBrz3HmA0AqGh4uytp54CgoKqvXYiInIcuwOftm3bAgBMJpPTOkOkFlhYBgXmx5kXNI9JicektASb51A7d00zr8zfW3NdjSAAX3wBPPPMlbTWnXcCb79tldZKiovA4eLzdqXciIiodmo9nX3v3r3Iy8uzKnS+7bbb6twp8l1qQYhl0GJ5nOVu6lrOXZNa7xWWmysuQigV/XfoAMyZA9x0k83DzfcBIyIi59C8gOHRo0fxr3/9C7m5udDpdJBertOJi/wbjUbH97KecAFD15uxYq+8YvLkIZ3tfp15OkoKgly2X1VpKfDf/wKzZ19Ja02ZIqa1goNVX8ZFCYmIasepe3WNHz8esbGxKC4uRmhoKPbs2YP169ejV69eWLt2bW37TAQAWJSTh/IqIxbl5Gl6nfn0bpetcSOltTp2BGbNEoOe4cOBffuAiROrDXoATlEnIqoPmgOfzZs34+WXX0azZs3g5+cHPz8/9OvXD5mZmXjyySed0UfyEPYsBFjTMZaztxzxnvVi924gJQUYORIoLASuugpYuRL49lsgxr5rISIi59Mc+BiNRjRu3BgA0KxZM5w8eRKAWPx84MABx/aOPIraVG/z4KSm6eCTh3TGvuk3W6W5zM9h/mdbixXaanOGrNwC3D1zFY6lPwZ06wasXw+EhAAzZoj1PYMHu09gRkREAGpR3Ny1a1fs3LkTsbGx6N27N2bOnImgoCAsWLAAcXFxzugjeQjLImDLDUTNa3C0pqEsA6bqgidbCxg6nCBgz5vzMed/7yHy/OVi5DvuEGdrXZ4BadlvprCIiFxPc+Dz4osv4sKFCwCAl19+Gbfeeiv69++Ppk2b4uuvv3Z4B8lzqM20Ml9rp7Yzq8yneveICVcET5aBlK11fBxqzx5g7Fg8s24dAOBCTCwafjAXuPlmq0NrPSOMiIicQvOsLltKSkoQHh4uz+zyVJzV5Vi1maWk9pqhczZYzdRSO7a2M8NqdO4c8NJLwLvvApcuiWmtyZPFNXpqKFwmIiLnceqsLlsiIiKcHvRkZmbi2muvRePGjdGiRQsMGzbMqqbo4sWLGDt2LJo2bYpGjRph+PDhKJIWjaN6V5tZSmo1QElxEQgJ9Fcs7icdm7Fkp6KGxuHr4QiCuFN6QoK4n9alS8CwYcDevWLgw6CHiMhjaA58Lly4gClTpqBPnz5o37494uLiFD/Osm7dOowdOxY5OTlYvXo1qqqqcNNNN8lpNwB46qmn8MMPP+Cbb77BunXrcPLkSdxxxx1O6xPZx1aBr1rRr60AJyu3AAs3/YXyKiNW7bkSyI5JiUeQvx/Kq4yKQmaHTmffu1fcKf2++4CTJ4H4eGDFCmDpUqBdu7qfn4iI6pXmIoiHHnoI69atw6hRoxAdHV1v6a2VK1cqfv/000/RokULbN26FQMGDIDBYMDHH3+ML7/8EjfccAMA4JNPPkGnTp2Qk5ODpKSkeumnL1NLPdkq8FUr+rU1WjN37RFUGm1nZI2XM7Xmhcy1rSNSOHcOePll4J13xBGeBg2AF14Qd1Rv0KBu5yYiIpfRHPhkZWVhxYoV6Nu3rzP6YzeDwQBATLMBwNatW1FVVYXU1FT5mISEBMTExGDz5s02A5+KigpUVFTIv5eWljq5195NLZgxL0yWgiNpRMdyVMbWflVjUuLlEZ1JaQmK9zOargRE0l5ddQp6BAFYsgTIyBBHeADg9tvFBQljY2t/XjfElaKJyBdpDnzCw8PlYMNVTCYTJkyYgL59+6Jr164AgMLCQgQFBaFJkyaKYyMjI1FYWGjzPJmZmXjppZec3V2foTaDyXwUJ+doCXLzxaDV1nYStkZ8qtu/a8qy3SgtF0d76jxtfN8+YNw44Ndfxd/j4sRtJ4YMqd353Byn2hORL9Jc4zN9+nRMnToVZWVlzuiPXcaOHYvdu3dj8eLFdTrPpEmTYDAY5J8TJ044qIe+Sa2Y2bzmpqb6G8vnq1sAMC0xGtH6EFQaTQgNCqh9Xc/588BzzwFXXy0GPQ0aiLO39uzx2qAHcHAtFBGRh7BrxKd79+6KWp7Dhw8jMjIS7dq1Q2BgoOLYbdu2ObaHFsaNG4cff/wR69evR+vWreX2qKgoVFZW4uzZs4pRn6KiIkRFRdk8V3BwMII5I6deSaM3UkBjmWapbtf1mnZt1zxqIQjAN9+Iaa38fLHtttvEuh4vS2vZ4pBaKCIiD2NX4DNs2DAnd6NmgiDgiSeewNKlS7F27VrEWtyYevbsicDAQGRnZ2P48OEAgAMHDiAvLw/Jycmu6DJBHLHJWLIT5VVGRfCSmbUfeSVlyMzaX+3Nt6YFAG3dvO2qXdm/X0xrZWeLv8fGimmtW2/VfpFEROQx7Ap8pk2b5ux+1Gjs2LH48ssvsWzZMjRu3Fiu29Hr9QgJCYFer8eDDz6IjIwMREREICwsDE888QSSk5M5o6sO6loAO3ftEZRXGaEDFAXL9qppVMJW/6odJTp/HnjlFXFriaoqcQ2eSZPEVFdIiOb+ERGRZ9Fc45Oeno7169c7oy/VmjdvHgwGA1JSUhAdHS3/mG+TMWvWLNx6660YPnw4BgwYgKioKPzvf/+r9756k5o2FZWordVjKK9CkL8OAqAoWJ6UloDEVnrFLC2182jtn83aFSmt1akT8PrrYtBz663iOj3TpjHoISLyEZq3rBg2bBh++ukntG3bFv/5z3+Qnp6OVq1aOat/9YpbVlizd8TH1pYSUluQvx+i9A0wKS3B5jnM30MKZMzPU10/7E5rPfEE8Msv4u+xseK2E0OH1uITISIid+PULSu+//575Ofn4/HHH8fXX3+Ndu3aIS0tDd9++y2qqqpq3WlyT/ZsOyGN7MREhCpmY0mjPZVGE/QhgarnMB+1UVu5OWPJTpsjT9X278IFMY119dVi0BMcLI7u7NnDoIeIyEfVaq+u5s2bIyMjAzt37sSWLVvQvn17jBo1Ci1btsRTTz2FQ4cOObqf5Mbmrj2CvJIyRXAjtYWFBFoFMpbMU1NqKzeXVxkREuhv39RrQQC++05Ma732mpjWuuUWMeD573+Z1iIi8mF12qS0oKAAq1evxurVq+Hv749bbrkFubm56Ny5M2bNmuWoPpKbs1VTI7WFBgWgvMqIRTl5qnt1mY/aVHeut+++puYC64MH8U+/gcCddwInToj7aS1bBvz4o7jPFhER+TTNgU9VVRW+++473HrrrWjbti2++eYbTJgwASdPnsTChQvxyy+/YMmSJXj55Zed0V9yEq1FxeZspZuktklpCQjy11ltJPrcd7uQm2/Ac9/tckj/ceGCuFN6165ovmkdKvwD8dXND4jFy7fdBtTTnnJEROTeNG9ZER0dDZPJhPvuuw+///47unXrZnXMwIEDrbaOIPfmiO0L1AqNL13eT+tMWaXcdv7iJcVjdf2otm+CIO6UPmGCOMIDoLjfDXhx4MP4113XM61FREQKmgOfWbNm4a677kKDanaobtKkCY4dO1anjlH9qmmhQHO2AhzLhQoBMWAxlFdB2ke0osoon2NQpxbI3leMQZ1aKM6ttkmpzb4dOiTO1lq1Svy9bVvg3XfR4rbbsIAjPEREZIPmwGfUqFHO6Ae5mJbtC2ytumy5UKE0ShMTEYpmjYJQWl6F9D7t5HMM79EaRaUVGN6jteLcdm1SWlYGvPoq8MYbQGUlEBQkLkA4aRIQGlrrz4CIiLxfnYqbyfvZW/szJiUeQf5+EACs2lMkT0sf3CUSf754Iw7OuAWTh3SWj5+ybDdy8w2Ysmy31XlUN86U0lqdOgEzZohBz803A7t3A9OnM+ghIqIaMfChatlaGXlwl0g5qJGkJUYjSn8l/SmN3KzaU2RzRedT58V6n7NmdT/SeWyuy3PokDgl/Y47gLw8ICZGDIJ++gm46ipHXjIREXkxBj5upi6zq5zB1giMrXQUoAyIpNcBsAqczP8c4F/DX8GyMmDKFKBrV2DlSjGtNXkysG8fMGwYZ2sREZEmDHzcjL17YzmDlrSWFAyZv2bVniJ5lMd8OrutdXmkcMVkUtkxRRDE9Xc6dxY3Fa2sBAYPFtNar7zCtBYREdUKAx83U22NixOpbQthKxAzT0fVJlBLS4zGQ/1jERLoryh4lh0+LG4gOmwYcPw4EBODbbM+xtBbJiPrYqM6XCUREfk6Bj5uxp69sZxBbVuImgIx8+dt1f6oBUaTh3TGvuk3KwqeUVYGTJ0KdOkC/PQTTAGBWHLTKPy85FdM8++A3JOlLhkJIyIi76F5Ojt5B8u1eMzXyrFcgbm6IMz8eSl4Mq/9sWt9IEEAfvgBGD8e+Osvse2mm/D4delYVaVH4u8FmtYZIiIiUqMTBEGlyML3aNnW3hOoraQMAEPnbEBuvgGJrfT44Yl+ms6hdt7q3k/tvM/E++P6uTOAFSvEJ9q0AWbNAu64A1m7C+0+HxER+S4t92+murxYdfU39tQSaan7AYBteWdwuPg8tuWdsTqPZdH0Rz/vReriuUi+PUUMegIDxQUI9+0Dhg8HdDqXpf2IiMh7MfDxYtUFN/YEFWp1P5FhwdBdfjS3KCdP3ond8jyKQOmHH/D5rNEYv+krBF2qBG68EcjNFVdjbtiw9hdMRERUAwY+XqyuIyZjUuLRuEEAyquM+G7b33L72gP/QLj8aK5v+6bQXX60PE9iKz2ebh8ADB0K3HYbQvPz8E+TFtj+5gJxr62OHQG43zpGRETkXRj4kKq0xGh59/TsfcVyu2D2aB6oFJVWQABQVFqhOI/fxXLcm/UJ+g9LAX78EQgIwLepI3D9f97H1KBOikUIXbmOEREReT8GPlStQZ1aQHf5UTK6bzuEBPpjdN92ikBF2p/LfGf1P9/7DJ1v6osRWf8H/8oKYNAgYNcuNHz7DcTHRlml4Vy1jhEREfkGBj4EwHaKKSu3AAeLzqNNRKjVLuoS80BFsZXFsWPAbbeh1xPpaHO2EIWNm2L7G/OB1avFTUZVsKCZiIiciYEPAQAys/YjN9+AzKz9ctvctUeQV1KGvJIyRfvCTX+hvMqIhZv+UgQqY1Li0aNFA7y9//JWEz/8AFNAAL4ddD92rc5B92celdNaaimtGSv2otOUlZixYm/9XDgREfkUBj4EACirvCQ/SsFHZFgw/P10iucBICwkUPEoab7+F7z/6r9x1dw3gYsXgRtugN+uXbjzly9wU+/2imPVUlpqM8OIiIgcgSs3kxUp+Nh4+DTCQwNx6nyl4vnpt3dVrqJ87BgwYQJ6LV8OAPgnrCmaL3gfuPtu1d3T1VaEHpkUg0U5eRiZFOPYiyIiIgJHfMiGkUkxCAn0x8ikGIQGibGx9AiY1eFcFQ5Mny6mtZYvxyU/f8y/7g4Mf/Jj4J57VIOe6qas29zDi4iIyEE44uNjzLeVACD/WQxsKhEaFIDJQzrLgUePmHB5xtbQORuQFBeBpdvz0W33Zsxc/xEiCk6IJx44EP/pPgq/BbZATOPQavtgXt/DImYiIqpPDHy8lFqAY1lULP15cJdILMrJU+ysDlxJSUl7e5UdPIIZq+Zj8KEc8YDoaODtt4F77sH9uwtx1o6NRLnhKBERuQo3KTXjTZuUmm9CCkD+s2XQIf05M2s/8krKEBMRivXPDbQ632tLtyPk3Xfw+KavEVRVgSo/f2y9fRSSFs4GGjeu12sjIiIyp+X+zREfL6UW4FgWFUt/Np+ubmXlSvz70YfR8h9x24rT1/bB1NTHcOt9qXYFPVp2eCciInImjviY8aYRH61spcYyOgSj0xvTEJWdBQA4HdYUJ154GQ9VtsepC1Vo1igIf754Y43nHjBzjdVokvmI1A9P9HPehRERkdfTcv/mrC4voLbqsq2ZU/ZsArpg9T70//ZD9BnaH1HZWbik88PSgfeg6Ymj6Pb8GJRe3r+rtLxK87kl3JqCiIhcgYGPF7C1CrLayshq7dLKzb+88zkWzX4Yz63/DMFVFTje9Vrc8fD72Pv0VOByFJ3eR9yrK71PO7vOPSktAYmt9JiUliC3cWsKIiJyBdb4eAFbs6TUZk6ptV91aCfWf/TUlYaoKOCttzDuVBvkniyFcLREfqpHTDhyjpagR0y4XedWW6yQiIiovrHGx4xP1vicOwe0aiU+Sp56Cvjvf4GwMJtFyLZqdoiIiFyFNT4EwI6am2efFdNXZkHPjsz3kPWfZzF04S7V15nv6+W0vhERETkBAx8vplZzg02bxO0k3nzzStvgwYDRiG4TxypeZ+sctraxcFjfiIiInIiBjxezmjl1/jwQHg707as8MC8PWLkS8POzep2t2Ve2ipXr3DciIqJ6wOJmL6YoKp44EXj9deUBn38OjBwJwHpBQfOFBqs9r0bm78P1e4iIqL5xxMcLVFsvk5MjprXMg54bbwSMRjnoAa5MZ7dcwdnRKSmmuIiIyJUY+Hi4rNwCZCzZaR1MXLgANGsGJCcrX3D8OPDzz3Jaq6ZzFxjKEeTvh6S4CIf0lykuIiJyJQY+Hm7u2iMorzIiJND/SjDxwgtAo0bA6dPycRlDnsLQ2b8BMTE2zzO4SyRCAv0Vu7PPXXsEp85XotJowtLt+Q7pLxcuJCIiV2Lg4+GkEZS3774GaWV5YlorM1N+fmeHHsjafgKHBt9R7SjLqj1FKK8yYtWeIsW5dZf/bLk9RV1xOjsREbkCi5vdmD07mKclRiMtLgyIjQX++Ufx3IPTluDO4eLoSlq31prfPy0xGg/1j8WinDyMTLI9UlRb5rU+HP0hIqL6whEfN2ZXIfCUKWJayzzo+fRTQBDw8X/vsjuoUJui3iMmHO1bNFJsT+GI0RrW+hARkStwxMeNqe19BQD4/Xegd29F06723XH1/j8Af3+H9cHWyIwjRmu4fxcREbkCR3zcmM1C4LIycQNRi6DnoWlfI/9/K2oMetRGa9Sms9sameFoDREReSqO+HiSadOAl19Wtn38MTB6ND6y8xRaR2tsjcxwtIaIiDwVAx9P8OefwLXXKtv69wfWrNGc1kqKi8Dh4vNW6/JMSktQT6sRERF5Caa63FlZGdCypXXQc+QIsH69atBTXfGxrWnrgP3r63AaOhEReTIGPu7qpZeAhg2BArMA46OPAEEA4uLkJluBSHWzwcoqL8mPtQliuOUEERF5MgY+7mbrVnERwv/+90pb377ApUvAgw9aHW4rEEmKi0CQvx8KDOVWQU1oUID8aP5ae4MgRxU2awm6OMpERESOwsDHXZSXA23aAL16KdsPHwY2bFBNa9kKRHKOlqDSaMKp85VWIzPm6/WYv9bekRxHbTmhZeSIo0xEROQoDHzcwfTpQGgo8PffV9oWLBDTWvHVj6zYCkTGpMSjWaMgBPnrqt1c1Py1SXERCAn0d9hmpDXRMnLE6fNEROQoDHxcaft2Ma01deqVtuRkrNyWh6EXO1Wb2qku/SMFQZVGwWpz0SnLdiM334Apy3Yr2nOOlqC8yoicoyV2vUddaRk54samRETkKAx8XOHiRaBdO6BHD2X7oUPApk14f8PxGlM7NaV/pE1FLTcXLS2/pHiU2BpVYYqJiIi8DQOf+jZjBhASAhw/fqVt3jwxrdW+PQD7Ujvmx9gamUnv0w4hgf5I79NO8boBHZpBd/nRnK1RFbX0F4uNiYjIU3EBw/qyYwfQvbuy7brrgI0bgQDl12DPysjmxwyds8FqNebJQzpj8pDOAJS7vBeVVkAAUFRaUWOXbaW/AO6sTkREnosjPvWhqso66Dl4ENiyxSroqQ21ESJpZEbah0sKfuwpFM7KLYChvAoxEaFWx7LYmIiIPJVXBj7vv/8+2rVrhwYNGqB37974/fffXduhgABg0CCpc2Ja66qrHHZ6teJfaWQGgOZAZe7aI8grKYM+JNDmXl0sNiYiIk/kdamur7/+GhkZGZg/fz569+6Nd955B4MHD8aBAwfQokUL13RKpwN++cXhp5VSWElxEcg5WoIxKfFyMGI+YjMpLUFu7/XKapw6XynP6jLfn0v6c1pitLy2D0d1iIjIm+gEQRBc3QlH6t27N6699lq89957AACTyYQ2bdrgiSeewMSJExXHVlRUoKLiSq1LaWkp2rRpA4PBgLCwsHrtd21ItT0hgf4orzIisZUePzzRT/GceRsAtH/hJ1wyCQjw06FTdJh8DACbxxMREbm70tJS6PV6u+7fXpXqqqysxNatW5Gamiq3+fn5ITU1FZs3b7Y6PjMzE3q9Xv5p06ZNfXa3zsakxCMmIhQNg/2tanHU6nBCgvzlR/NjtNTtcFYXERF5Kq9KdZ06dQpGoxGRkZGK9sjISOzfv9/q+EmTJiEjI0P+XRrx8RRpidFyLU5IoNHqOS0zw6Tf7cFZXURE5Km8asRHq+DgYISFhSl+PM2YlHg51WXPQoMVVUbFo0TLKA5ndRERkafyqhGfZs2awd/fH0VFRYr2oqIiREVFuahXjmO+Ho800iI9WhYiqxU+p/dph0U5eRiZFKM4t5ZRHHtGk4iIiNyRV434BAUFoWfPnsjOzpbbTCYTsrOzkZyc7MKeOUZtdjRflJNn12u4SjMREfkCrwp8ACAjIwMffvghFi5ciH379uHxxx/HhQsX8J///MfVXasztRSTtEBhZtZ+q2NHJsUoXrNw018orzJi4aa/FOdYtacI5VVGrNqjHC3jfl1ERORNvCrVBQD33HMP/vnnH0ydOhWFhYXo1q0bVq5caVXw7M5spbQAZYrJ/Jjq9IgJl7euAIDgQH9UGi8hONDfrr5wPR8iIvImXjfiAwDjxo3D8ePHUVFRgS1btqB3796u7pIm9oyymB8zuEskQgL9MbhLZI3nCA8NUjxKJqUlILGVHpPSEhTtXKWZiIi8iVcGPp6upr23snILFMfY2kxU7RzODHBYD0RERO7O61Jd3kBt1pT5KI5lkJKZtR+G8ipk5Ra4bHSG6/sQEZG744iPB1EbxUlLjIY+JBB5JWVyastWwXN17Y6gNjOMiIjIXXDEx4NUt36OOxQh20q5ERERuROO+LiYo+piLGt01Gp5bBVCOwpXdCYiInfHER8Xc1ZdjNrokDNHZbiiMxERuTuO+LhYfY+ScFSGiIh8GUd8XIyjJERERPWHIz71xFVr3Fi+L7egICIiX8bAp544M+CoLqiyfF+muoiIyJcx8Kknzgw4qguqLN/XfPYXV1omIiJfoxMEQXB1J9xFaWkp9Ho9DAYDwsLCXN0du1luWGprg1Nbhs7ZgNx8AxJb6fHDE/1qPDdrkYiIyB1puX9zxMcLmI/i1JRSU9vvSw1rgoiIyJtwVpcXycotgKG8CjERoarBTHX7fdniDitCExEROQoDHy8yd+0R5JWUIbGV3mFbW3C6PREReROmurxIdakrKcUFwK6RHiIiIm/EER8vUt3ojLO2xiAiIvIkHPHxAjNW7EWnKSsxY8Ve1WO4fg8REREDH6+wKCcP5VVGLMrJU7Sbz+Cy3L2diIjIFzHw8QJ92zeF7vKjOU5FJyIiUmLg4wWKSisgXH6UzFixFwcKz6FZoyCmt4iIiC5j4ONCjtoywlb9zqKcPFQaTbhQYWR6i4iI6DIGPi7kqFSUrfqdkUkxCAn0x8ikmLp2k4iIyGtwry4z9b1XF/fBIiIiqjst92+u4+NCXBWZiIiofjHVRURERD6DgQ8RERH5DAY+bshRs72IiIhIiYGPG+LCg0RERM7BwMcNad1XiyNERERE9uGsLjekdbYXd14nIiKyD0d83JDWERzuvE5ERGQfBj5uSGuNz7a8MzhcfB7b8s44uWdERESejYGPG9I6grMoJw/lVUYsyslzcs+IiIg8G2t83JDWGp+RSTFYlJPHfbmIiIhqwL26zNT3Xl1acW8vIiIia1ru30x1eRCu70NERFQ3DHzcTHUzujh7i4iIqG4Y+LgZjuoQERE5DwMfN1PdqA6DIiIiorrhrC43U92MrjEp8XJxMxEREWnHwMeDaJ3mTkREREpMdREREZHPYOBDREREPoOBDxEREfkMBj5ERETkMxj4EBERkc9g4ENEREQ+g4EPERER+QwGPkREROQzGPgQERGRz2DgQ0RERD6DgY8bysotwNA5G5CVW+DqrhAREXkVBj4uZivI4S7sREREzsHAx8VsBTljUuKR2ErPXdiJiIgcjLuzu9iYlHjMXXtEEeRwF3YiIiLnYODjYgxyiIiI6o9HpLr++usvPPjgg4iNjUVISAji4+Mxbdo0VFZWKo7btWsX+vfvjwYNGqBNmzaYOXOmi3pMRERE7sgjRnz2798Pk8mEDz74AO3bt8fu3bvx8MMP48KFC3jzzTcBAKWlpbjpppuQmpqK+fPnIzc3F6NHj0aTJk3wyCOPuPgKiIiIyB3oBEEQXN2J2njjjTcwb948HD16FAAwb948TJ48GYWFhQgKCgIATJw4Ed9//z32799v1zlLS0uh1+thMBgQFhbmtL4TERGR42i5f3tEqssWg8GAiIgI+ffNmzdjwIABctADAIMHD8aBAwdw5swZm+eoqKhAaWmp4oeIiIi8l0cGPocPH8acOXPw6KOPym2FhYWIjIxUHCf9XlhYaPM8mZmZ0Ov18k+bNm2c12kiIiJyOZcGPhMnToROp6v2xzJNlZ+fj5tvvhl33XUXHn744Tq9/6RJk2AwGOSfEydO1Ol8RERE5N5cWtz89NNP44EHHqj2mLi4OPnPJ0+exMCBA9GnTx8sWLBAcVxUVBSKiooUbdLvUVFRNs8dHByM4ODgWvSciIiIPJFLA5/mzZujefPmdh2bn5+PgQMHomfPnvjkk0/g56ccrEpOTsbkyZNRVVWFwMBAAMDq1avRsWNHhIeHO7zvRERE5Hk8osYnPz8fKSkpiImJwZtvvol//vkHhYWFitqd+++/H0FBQXjwwQexZ88efP3113j33XeRkZHhwp4TERGRO/GIdXxWr16Nw4cP4/Dhw2jdurXiOWk2vl6vx88//4yxY8eiZ8+eaNasGaZOnco1fIiIiEjmsev4OAPX8SEiIvI8PrGODxEREZFWDHyIiIjIZ3hEjU99kbJ+XMGZiIjIc0j3bXuqdxj4mDl37hwAcAVnIiIiD3Tu3Dno9fpqj2FxsxmTyYSTJ0+icePG0Ol0tTpHaWkp2rRpgxMnTnhtgTSv0Xv4wnXyGr0Dr9E7OOsaBUHAuXPn0LJlS6t1/ixxxMeMn5+f1XT52goLC/Pav7gSXqP38IXr5DV6B16jd3DGNdY00iNhcTMRERH5DAY+RERE5DMY+DhYcHAwpk2b5tWbn/IavYcvXCev0TvwGr2DO1wji5uJiIjIZ3DEh4iIiHwGAx8iIiLyGQx8iIiIyGcw8CEiIiKfwcDHwd5//320a9cODRo0QO/evfH777+7uku1tn79egwdOhQtW7aETqfD999/r3heEARMnToV0dHRCAkJQWpqKg4dOuSaztZSZmYmrr32WjRu3BgtWrTAsGHDcODAAcUxFy9exNixY9G0aVM0atQIw4cPR1FRkYt6rN28efNw9dVXywuGJScnIysrS37e06/Pltdeew06nQ4TJkyQ2zz9Ov/73/9Cp9MpfhISEuTnPf36JPn5+Rg5ciSaNm2KkJAQJCYm4s8//5Sf9/T/77Rr187qe9TpdBg7diwA7/gejUYjpkyZgtjYWISEhCA+Ph7Tp09X7KPl0u9RIIdZvHixEBQUJPzf//2fsGfPHuHhhx8WmjRpIhQVFbm6a7Xy008/CZMnTxb+97//CQCEpUuXKp5/7bXXBL1eL3z//ffCzp07hdtuu02IjY0VysvLXdPhWhg8eLDwySefCLt37xZ27Ngh3HLLLUJMTIxw/vx5+ZjHHntMaNOmjZCdnS38+eefQlJSktCnTx8X9lqb5cuXCytWrBAOHjwoHDhwQHjhhReEwMBAYffu3YIgeP71Wfr999+Fdu3aCVdffbUwfvx4ud3Tr3PatGlCly5dhIKCAvnnn3/+kZ/39OsTBEEoKSkR2rZtKzzwwAPCli1bhKNHjwqrVq0SDh8+LB/j6f/fKS4uVnyHq1evFgAIa9asEQTBO77HGTNmCE2bNhV+/PFH4dixY8I333wjNGrUSHj33XflY1z5PTLwcaDrrrtOGDt2rPy70WgUWrZsKWRmZrqwV45hGfiYTCYhKipKeOONN+S2s2fPCsHBwcJXX33lgh46RnFxsQBAWLdunSAI4jUFBgYK33zzjXzMvn37BADC5s2bXdXNOgsPDxc++ugjr7u+c+fOCVdddZWwevVq4frrr5cDH2+4zmnTpgnXXHONzee84foEQRCef/55oV+/fqrPe+P/d8aPHy/Ex8cLJpPJa77HIUOGCKNHj1a03XHHHcKIESMEQXD998hUl4NUVlZi69atSE1Nldv8/PyQmpqKzZs3u7BnznHs2DEUFhYqrlev16N3794efb0GgwEAEBERAQDYunUrqqqqFNeZkJCAmJgYj7xOo9GIxYsX48KFC0hOTva66xs7diyGDBmiuB7Ae77HQ4cOoWXLloiLi8OIESOQl5cHwHuub/ny5ejVqxfuuusutGjRAt27d8eHH34oP+9t/9+prKzEokWLMHr0aOh0Oq/5Hvv06YPs7GwcPHgQALBz505s2LABaWlpAFz/PXKTUgc5deoUjEYjIiMjFe2RkZHYv3+/i3rlPIWFhQBg83ql5zyNyWTChAkT0LdvX3Tt2hWAeJ1BQUFo0qSJ4lhPu87c3FwkJyfj4sWLaNSoEZYuXYrOnTtjx44dXnF9ALB48WJs27YNf/zxh9Vz3vA99u7dG59++ik6duyIgoICvPTSS+jfvz92797tFdcHAEePHsW8efOQkZGBF154AX/88QeefPJJBAUFIT093ev+v/P999/j7NmzeOCBBwB4x99TAJg4cSJKS0uRkJAAf39/GI1GzJgxAyNGjADg+vsHAx+iy8aOHYvdu3djw4YNru6Kw3Xs2BE7duyAwWDAt99+i/T0dKxbt87V3XKYEydOYPz48Vi9ejUaNGjg6u44hfSvZQC4+uqr0bt3b7Rt2xZLlixBSEiIC3vmOCaTCb169cKrr74KAOjevTt2796N+fPnIz093cW9c7yPP/4YaWlpaNmypau74lBLlizBF198gS+//BJdunTBjh07MGHCBLRs2dItvkemuhykWbNm8Pf3t6q+LyoqQlRUlIt65TzSNXnL9Y4bNw4//vgj1qxZg9atW8vtUVFRqKysxNmzZxXHe9p1BgUFoX379ujZsycyMzNxzTXX4N133/Wa69u6dSuKi4vRo0cPBAQEICAgAOvWrcPs2bMREBCAyMhIr7hOc02aNEGHDh1w+PBhr/keo6Oj0blzZ0Vbp06d5JSeN/1/5/jx4/jll1/w0EMPyW3e8j0+++yzmDhxIu69914kJiZi1KhReOqpp5CZmQnA9d8jAx8HCQoKQs+ePZGdnS23mUwmZGdnIzk52YU9c47Y2FhERUUprre0tBRbtmzxqOsVBAHjxo3D0qVL8euvvyI2NlbxfM+ePREYGKi4zgMHDiAvL8+jrtOSyWRCRUWF11zfoEGDkJubix07dsg/vXr1wogRI+Q/e8N1mjt//jyOHDmC6Ohor/ke+/bta7WcxMGDB9G2bVsA3vP/HQD45JNP0KJFCwwZMkRu85bvsaysDH5+yvDC398fJpMJgBt8j04vn/YhixcvFoKDg4VPP/1U2Lt3r/DII48ITZo0EQoLC13dtVo5d+6csH37dmH79u0CAOHtt98Wtm/fLhw/flwQBHE6YpMmTYRly5YJu3btEm6//XaPmlYqCILw+OOPC3q9Xli7dq1iimlZWZl8zGOPPSbExMQIv/76q/Dnn38KycnJQnJysgt7rc3EiROFdevWCceOHRN27dolTJw4UdDpdMLPP/8sCILnX58a81ldguD51/n0008La9euFY4dOyZs3LhRSE1NFZo1ayYUFxcLguD51ycI4lIEAQEBwowZM4RDhw4JX3zxhRAaGiosWrRIPsYb/r9jNBqFmJgY4fnnn7d6zhu+x/T0dKFVq1bydPb//e9/QrNmzYTnnntOPsaV3yMDHwebM2eOEBMTIwQFBQnXXXedkJOT4+ou1dqaNWsEAFY/6enpgiCIUxKnTJkiREZGCsHBwcKgQYOEAwcOuLbTGtm6PgDCJ598Ih9TXl4ujBkzRggPDxdCQ0OFf/3rX0JBQYHrOq3R6NGjhbZt2wpBQUFC8+bNhUGDBslBjyB4/vWpsQx8PP0677nnHiE6OloICgoSWrVqJdxzzz2K9W08/fokP/zwg9C1a1chODhYSEhIEBYsWKB43hv+v7Nq1SoBgM1+e8P3WFpaKowfP16IiYkRGjRoIMTFxQmTJ08WKioq5GNc+T3qBMFsKUUiIiIiL8YaHyIiIvIZDHyIiIjIZzDwISIiIp/BwIeIiIh8BgMfIiIi8hkMfIiIiMhnMPAhIiIin8HAh4iIiHwGAx8iIiLyGQx8iIiIyGcw8CEiIiKfwcCHiLzaZ599hqZNm6KiokLRPmzYMIwaNcpFvSIiV2HgQ0Re7a677oLRaMTy5cvltuLiYqxYsQKjR492Yc+IyBUY+BCRVwsJCcH999+PTz75RG5btGgRYmJikJKS4rqOEZFLMPAhIq/38MMP4+eff0Z+fj4A4NNPP8UDDzwAnU7n4p4RUX3TCYIguLoTRETO1rNnT9x555246aabcN111+Gvv/5CmzZtXN0tIqpnAa7uABFRfXjooYfwzjvvID8/H6mpqQx6iHwUR3yIyCcYDAa0bNkSly5dwmeffYZ77rnH1V0iIhdgjQ8R+QS9Xo/hw4ejUaNGGDZsmKu7Q0QuwsCHiHxGfn4+RowYgeDgYFd3hYhchKkuIvJ6Z86cwdq1a3HnnXdi79696Nixo6u7REQuwuJmIvJ63bt3x5kzZ/D6668z6CHycRzxISIiIp/BGh8iIiLyGQx8iIiIyGcw8CEiIiKfwcCHiIiIfAYDHyIiIvIZDHyIiIjIZzDwISIiIp/BwIeIiIh8xv8D7139YZbvyuMAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "yhatc = lmmodc.predict(XX)\n",
    "plt.scatter(y,yhatc,s=.8)\n",
    "plt.plot(y,y,c='red')\n",
    "plt.xlabel('y'); plt.ylabel('yhat with trim')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['550' '550' '550' '550' '550' '500']\n",
      "['darkgrey' 'darkgrey' 'darkgrey' 'darkgrey' 'darkgrey' 'purple']\n"
     ]
    }
   ],
   "source": [
    "cvec = cdn['trim'].to_numpy()\n",
    "print(cvec[:6])\n",
    "cvec[cvec=='430'] = 'black'\n",
    "cvec[cvec=='500'] = 'purple'\n",
    "cvec[cvec=='550'] = 'darkgrey'\n",
    "cvec[cvec=='other'] = 'red'\n",
    "print(cvec[:6])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x73611a757fd0>"
      ]
     },
     "execution_count": 117,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(yhat,yhatc,s=.8,c=cvec) # the c argument accepts a vector of colors!!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's get R-squared."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.8518"
      ]
     },
     "execution_count": 118,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp = np.corrcoef(y,yhatc)[0,1]\n",
    "np.round(temp**2,4) # R-squared"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "R-squared was .83 with just mileage and year.\n",
    "It is not the much bigger with with trim, but from the plot it looks like trim is useful."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Regression In Matrix Notation\n",
    "\n",
    "Let's write our multiple regression model using vector/matrix notation and use \n",
    "basic matrix operations to check the predicted and fitted values.\n",
    "\n",
    "The general multiple regression model is written:\n",
    "\n",
    "$$\n",
    "Y_i = \\beta_0 + \\beta_1  x_{i1} + \\beta_2  x_{i2} + \\ldots  \\beta_p  x_{ip} + \\epsilon_i, \\; i=1,2,\\ldots,n,\n",
    "$$\n",
    "\n",
    "where $i$ indexes observations and $x_{ij}$ is the value of the $j^{th}$ $x$ in the \n",
    "$i^{th}$ observation.\n",
    "\n",
    "If we let  \n",
    "\n",
    "\\begin{equation}\n",
    "x_i = \\left[\n",
    "\\begin{array}{c}\n",
    "1 \\\\\n",
    "x_{i1} \\\\\n",
    "x_{i2} \\\\\n",
    "\\vdots \\\\\n",
    "x_{ip}\n",
    "\\end{array}\n",
    "\\right], \\;\n",
    "X= \\left[\n",
    "\\begin{array}{c}\n",
    "x_1' \\\\\n",
    "x_2' \\\\\n",
    "\\vdots \\\\\n",
    "x_n'\n",
    "\\end{array}\n",
    "\\right], \\;\\;\n",
    "y = \\left[\n",
    "\\begin{array}{c}\n",
    "y_1 \\\\\n",
    "y_2 \\\\\n",
    "\\vdots \\\\\n",
    "y_n\n",
    "\\end{array}\n",
    "\\right], \\;\\;\n",
    "\\epsilon = \\left[\n",
    "\\begin{array}{c}\n",
    "\\epsilon_1 \\\\\n",
    "\\epsilon_2 \\\\\n",
    "\\vdots \\\\\n",
    "\\epsilon_n\n",
    "\\end{array}\n",
    "\\right], \\;\\;\n",
    "\\beta = \\left[\n",
    "\\begin{array}{c}\n",
    "\\beta_0 \\\\\n",
    "\\beta_1 \\\\\n",
    "\\beta_2 \\\\\n",
    "\\vdots \\\\\n",
    "\\beta_p\n",
    "\\end{array}\n",
    "\\right]\n",
    "\\end{equation},\n",
    "\n",
    "then we can write the model in matrix form:\n",
    "\n",
    "$$\n",
    "y = X \\beta + \\epsilon.\n",
    "$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In our data, the first three rows of $X$ are"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1.00000e+00, 3.68580e+01, 2.00800e+03],\n",
       "       [1.00000e+00, 4.68830e+01, 2.01200e+03],\n",
       "       [1.00000e+00, 1.08759e+02, 2.00700e+03]])"
      ]
     },
     "execution_count": 119,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X[0:3,:]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Which correspond to the the first three rows of our data frame cd:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "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>mileage</th>\n",
       "      <th>year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>36.858</td>\n",
       "      <td>2008</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>46.883</td>\n",
       "      <td>2012</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>108.759</td>\n",
       "      <td>2007</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   mileage  year\n",
       "0   36.858  2008\n",
       "1   46.883  2012\n",
       "2  108.759  2007"
      ]
     },
     "execution_count": 120,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cd.iloc[0:3,1:3] "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Given our estimates:\n",
    "\n",
    "\\begin{equation}\n",
    "\\hat{\\beta} = \\left[\n",
    "\\begin{array}{c}\n",
    "\\hat{\\beta}_0 \\\\\n",
    "\\hat{\\beta}_1 \\\\\n",
    "\\vdots \\\\\n",
    "\\hat{\\beta}_p\n",
    "\\end{array}\n",
    "\\right]\n",
    "\\end{equation}\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can get fitted values or predictions by matrix multiplication:\n",
    "$$\n",
    "\\hat{y} = X \\, \\hat{\\beta}, \\;\\; \\mbox{or}, \\;\\; \\hat{y}_p = X_p \\, \\hat{\\beta}.\n",
    "$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In our example,  \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(3, 1)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "array([[-5.36548987e+03],\n",
       "       [-1.53721903e-01],\n",
       "       [ 2.69434954e+00]])"
      ]
     },
     "execution_count": 121,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bhat = np.hstack([lmmod.intercept_,lmmod.coef_])[:,np.newaxis]\n",
    "print(bhat.shape)\n",
    "bhat"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "So we can get our predictions by multiplying $X_p$ times $\\hat{\\beta}$.  \n",
    "\n",
    "But first we have to add the column of ones:  \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Xp:\n",
      " [[  40. 2010.]\n",
      " [ 100. 2004.]]\n",
      "Xpp:\n",
      " [[1.000e+00 4.000e+01 2.010e+03]\n",
      " [1.000e+00 1.000e+02 2.004e+03]]\n"
     ]
    }
   ],
   "source": [
    "Xpp = np.hstack([np.ones((2,1)),Xp])\n",
    "print(\"Xp:\\n\",Xp)\n",
    "print(\"Xpp:\\n\",Xpp)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we can matrix multiply $X_{pp}$ times $\\hat{\\beta}$:  \n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[44.00383414],\n",
       "       [18.61442272]])"
      ]
     },
     "execution_count": 123,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yhatp = Xpp @ bhat # Xpp * bhat, matrix multiplication\n",
    "yhatp"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is the same as what we got using the predict method on the lmmod object.  \n",
    "\n",
    "Let's get the in-sample fitted values by multiplying $X \\hat{\\beta}$:  \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[39.09812927]\n",
      " [48.33446537]\n",
      " [25.3510212 ]]\n",
      "[39.09812927 48.33446537 25.3510212 ]\n"
     ]
    }
   ],
   "source": [
    "yhatm = X @ bhat\n",
    "print(yhatm[0:3,:])\n",
    "print(yhat[0:3]) #got these ones using the predict method"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1000,)"
      ]
     },
     "execution_count": 125,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dyhat = yhatm.flatten() - yhat\n",
    "dyhat.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "-9.094947017729283e-16"
      ]
     },
     "execution_count": 126,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dyhat.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "8.263534319404751e-28"
      ]
     },
     "execution_count": 127,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dyhat.var()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```dyhat``` has 0 mean and variance, so it must be all zeros.  "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Just for fun we can plot yhat vs yhatm:  \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(yhat,yhatm)\n",
    "plt.scatter(yhat,yhat,color='red',s=.5)\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.10.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}