{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Linear Regression (선형회귀)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 회귀\n",
    "  - 회귀: '다시 본디 상태로 되돌아 온다'는 뜻\n",
    "  - 영국의 유전학자 갈톤 (F. Galton, 1822 ~ 1891)에 의해서 회귀라는 용어가 등장\n",
    "    - 그의 유전에 관한 논문 'Family Likeness in Stature (1886)'에서 처음 회귀에 대하여 정의함\n",
    "      - 일반적으로 키가 큰 부모에게서 키 큰 자녀가, 키가 작은 부모에게서 키 작은 자녀가 태어난다. 그렇지만 자녀들의 평균 키는 전체 인구의 평균 키로 회귀하는 경향이 있다. \n",
    "      - 즉, 키가 큰 부모이든 작은 부모이든 그들에게서 태어난 자녀들의 평균 키는 전체 평균 키 수준에 접근하는 현상으로 나타난다\n",
    "      - 이러한 현상을 '보편적 회귀법칙 (law of universal regression)' 이라 한다.\n",
    "  - 보편적 회귀법칙은 피어슨 (K. Pearson, 1857 ~ 1936) 에 의해서 체계적인 회귀분석 이론으로 정립되었음 \n",
    "    - 피어슨은, 1,078 개 가족 구성원 자료를 수집하여 아버지 키와 아들 키 사이에 존재하는 회귀법칙을 규명하였다. \n",
    "    - 피어슨의 회귀법칙 결론\n",
    "      - 아버지 키가 비교적 큰 집단에서 태어난 아들들의 평균 키는 그 아버지 평균 키보다는 작다.\n",
    "      - 아버지 키가 비교적 작은 집단에서 태어난 아들들의 평균 키는 그 아버지 평균 키보다 크다.\n",
    "      - 따라서 아들의 키는 전체 평균 키 수준을 향하여 회귀하는 현상을 나타냄.\n",
    "\n",
    "<!-- more -->\n",
    "\n",
    "- 변수의 종류\n",
    "  - 결과 변수 (Outcome variable): 반응 변수 (Response Variable), 종속 변수 (Dependent Variable), 분류명 (Classified Name)\n",
    "  - 예측 변수 (Predictor): 설명 변수 (Explanatory Variable), 독립 변수 (Independent Variable), 특징 (Feature)\n",
    "  \n",
    "<!-- more -->\n",
    "\n",
    "- 회귀분석\n",
    "  - 한 종속변수가 하나 이상의 독립변수에 의해 어떠한 영향을 받고 또한 어떠한 관계로 나타나는지 분석하는 기법\n",
    "  - 결과 변수의 값을 다른 예측 변수들로 부터 설명하고 예측하는 것\n",
    "\n",
    "<!-- more -->\n",
    "\n",
    "- 회귀분석의 종류\n",
    "  - 선형회귀 (Linear Regression)\n",
    "    - 여러 개의 예측 변수와 하나의 결과 변수 사이에 선형 관계 (Linear Relationship)가 존재한다는 가정하에 그러한 변수 사이의 선형 구조를 *모형화* 하는 것\n",
    "    - **선형 관계 (Linear Relationship)**\n",
    "      - 예측 변수 $X1$에 대해서 결과 변수 $Y1$이 결정되고, 예측 변수 $X2$에 대해서는 결과 변수 $Y2$가 결정된다고 할 때, $aX1 + bX2$에 대해서는 결과 변수가 $aY1 + bY2$가 될 때 예측 변수와 결과 변수 사이에는 선형 관계가 있다고 말한다.\n",
    "  - 비선형회귀 (Nonlinear Regression)\n",
    "    - 종속변수와 독립변수 간의 임의적 관계로서 모형화\n",
    "\n",
    "<!-- more -->\n",
    "\n",
    "- 선형회귀분석 모형 (Hypothesis Function)\n",
    "  - 회귀 분석 알고리즘 및 훈련 데이터를 통하여 임의의 예측 함수 h를 생성\n",
    "  - 예측 함수 h를 hypothesis 라고 부름\n",
    "  - 예측 함수 h에 새로운 예측 변수를 넣으면 결과 변수를 출력함 (결과를 예측함)\n",
    "  \n",
    "<!-- more -->\n",
    "\n",
    "- Hypothesis Function 정의\n",
    "  - $n$개의 속성($n$개의 예측 변수)을 지닌 훈련 데이터 $n$-벡터 $x^i=\\{x_{1}, x_{2}, ..., x_{n}\\}$가 총 $m$ (즉, $1 \\le i \\le m$)개 주어지고,\n",
    "  - 각 $x^i$ 벡터마다 연관된 실수 값 $y^i$ (결과 변수)이 주어질 때,\n",
    "  - 임의의 $n$-벡터 $x^i =\\{x_1, x_2,...,x_n\\}$에 대해 Hypothesis Function $h_{\\theta}(x^i)$ 는 다음과 같이 정의된다.\n",
    "    $$h_{\\theta}(x^i) = \\theta_0 + \\theta_1 x_1 + \\theta_2 x_2 + ... + \\theta_n x_n$$\n",
    "    \n",
    "    - 위 식에서 $\\theta=\\{\\theta_0, \\theta_1, \\theta_2, ..., \\theta_n\\}$는 계수 벡터(Coefficient Vector)이다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 계수 벡터 $\\theta$를 구하는 수학적 모델\n",
    "  - 주어진 통계적 수치(훈련 데이터)들에 대해 다음 비용 함수 (Cost Function) $J(\\theta)$를 구한다. \n",
    "  \n",
    "  $$J(\\theta) = \\dfrac{1}{2m} \\sum_{i=1}^m \\big( h_\\theta(x^i) - y^i \\big)^2$$\n",
    "  \n",
    "    - 비용 함수 $J(\\theta)$를 Square Error Function 이라고도 칭함\n",
    "  - 선형회귀분석의 목적: 비용 함수 $J(\\theta)$를 최소로 만드는 $\\hat{\\theta}$ 벡터를 찾기\n",
    "  \n",
    "  $$\\newcommand{\\argmin}{\\arg\\!\\min} \\hat{\\theta} = \\argmin_\\theta J(\\theta)$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 비용 함수 $J(\\theta)$를 최소로 만드는 $\\hat{\\theta}$ 벡터를 구하는 방법\n",
    "  - 대표적인 방법: Gradient Descent"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Univariate Linear Regression (단일 변수 선형 회귀분석)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- $m$개의 통계치 $x^i$ (즉, $1 \\le i \\le m$)와 이와 연관된 실수 값 $y^i$에 대하여 새로운 통계 수치 $x$와 연관된 실수 값 $y$를 예측하기 위해 다음과 같은 모형 (Hyporthesis Function)을 고려함\n",
    "    $$h_{\\theta_0, \\theta_1}(x^i) = \\theta_0 + \\theta_1 \\cdot x^i$$    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 최적의 $h$ 함수를 위한 $\\theta_0$와 $\\theta_1$을 구하기 위하여 다음 비용 함수 $J(\\theta_0, \\theta_1)$를 고려\n",
    "    $$J(\\theta_0, \\theta_1) = \\dfrac{1}{2m}\\sum_{i=1}^m \\big( \\theta_0 + \\theta_1\\cdot x^i - y^i \\big)^2$$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1) 필요한 모듈을 지역 이름 공간에 불러오기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from scipy import stats\n",
    "from pandas import Series, DataFrame\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import scipy\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2) 통계 수치 데이터 정의 및 Pandas를 활용한 데이터프레임 정의\n",
    "- 최고 기온과 에어콘 판매 대수 데이터 (다소 극단적인 예) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "data = {\n",
    "    'Temperature': [26, 27, 28, 29, 30, 31, 32, 33],\n",
    "    'Number of Sells': [270, 280, 290, 300, 310, 320, 330, 340]\n",
    "}\n",
    "df = pd.DataFrame(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Number of Sells</th>\n",
       "      <th>Temperature</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>270</td>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>280</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>290</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>300</td>\n",
       "      <td>29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>310</td>\n",
       "      <td>30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>320</td>\n",
       "      <td>31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>330</td>\n",
       "      <td>32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>340</td>\n",
       "      <td>33</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Number of Sells  Temperature\n",
       "0              270           26\n",
       "1              280           27\n",
       "2              290           28\n",
       "3              300           29\n",
       "4              310           30\n",
       "5              320           31\n",
       "6              330           32\n",
       "7              340           33"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([26, 27, 28, 29, 30, 31, 32, 33])"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['Temperature'].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([270, 280, 290, 300, 310, 320, 330, 340])"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['Number of Sells'].values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3) 통계 수치에 대한 Scatter Plot 생성"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x11672dd90>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYkAAAEPCAYAAAC3NDh4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3X+YXVV97/H3J8lM7kEMqIyCiT0BQQcsI6Qm2tofZ2xD\n4d5boNYitfda69ze0ijwVGslFMxY7SNYsUXbtFYHm/bBSZBapAgBInP06pUkQiLRJNxYPTHGxp5q\nDWKOmJjv/WOvgcOYPTknmT3nTPJ5Pc95svfae6/9nWQy31lr7b2WIgIzM7NDmdXpAMzMrHs5SZiZ\nWS4nCTMzy+UkYWZmuZwkzMwsl5OEmZnlKjRJSJorab2kTZK2SFqRyldI+qakh9PnwqZrlkvaIWmb\npAuKjM/MzCanot+TkHRCROyTNBv4PHAVcBHw/Yh4/4RzzwY+BiwGFgDrgLPCL3OYmXVE4d1NEbEv\nbc4F5gDjP/B1iNMvAVZHxIGIqAE7gCVFx2hmZodWeJKQNEvSJmAPcH9EbEyH3ixps6SPSDoplc0H\ndjVdvjuVmZlZB0xHS+JgRJxP1n20RNI5wErgjIg4jyx53FR0HGZm1r4503WjiHhMUhW4cMJYxIeB\nf0nbu4EXNB1bkMqeRpLHKMzMjkBEHKqrP1fRTzedMt6VJKkELAW2Szq16bRXA19O23cCl0vqlXQ6\ncCaw4VB1R0TXfVasWNHxGByTYzoe43JMrX2ORNEtidOAVZJmkSWkNRFxt6R/kHQecBCoAb8PEBFb\nJd0GbAX2A8viSL8yMzM7aoUmiYjYAiw6RPnrJ7nmPcB7iozLzMxa4zeup1ClUul0CD/BMbXGMbWu\nG+NyTMUp/GW6IkhyL5SZWZskEd00cG1mZjObk4SZmeVykjAzs1xOEmZmlstJwszMcjlJmJlZLicJ\nMzPL5SRhZma5nCTMzCyXk4SZmeVykjAzs1xOEmZmlstJwszMcjlJmJlZrqKXL50rab2kTZK2SFox\n4fhbJR2U9OymsuWSdkjaJumCIuMzM7PJFb0y3ROSBiNin6TZwOcl3RMRGyQtIFvzeuf4+ZLOBi4D\nzgYWAOskneXFI8xsJqnX69RqNRYuXEhfX1+nwzkqhXc3RcS+tDmXLCmN/8D/C+BtE06/BFgdEQci\nogbsAJYUHaOZ2VQZHV1DudzP0qVXUC73Mzq6ptMhHZXCk4SkWZI2AXuA+yNio6SLgV1pDexm84Fd\nTfu7U5mZWder1+sMDS2j0Rhj796HaDTGGBpaRr1e73RoR6zQ7iaAiDgInC9pHvDPks4FriXrajpi\nw8PDT25XKpVjZj1ZM5u5arUavb0LaTQGUskAPT1larVaR7qdqtUq1Wr1qOqY1jWuJV1P1t30ZmAf\nILKxh91k3UpvBIiIG9L5a4EVEbF+Qj0epjCzrlOv1ymX+2k0xoAB4BFKpUF27tzeFWMTXbfGtaRT\nJJ2UtktkrYeHI+LUiDgjIk4HvgmcHxH/DtwJvFZSr6TTgTOBDUXGaGY2Vfr6+hgZWUmpNMi8eYso\nlQYZGVnZFQniSBXakkhdS6vIktEsYE1E/NmEc74GvCwivpv2lwNDwH7g6oi47xD1uiVhZl2rW59u\nOpKWxLR2N00VJwkzs/Z1XXeTmZnNbE4SZmaWy0nCzMxyOUmYmVkuJwkzM8vlJGFmZrmcJMzMLJeT\nhJmZ5XKSMDOzXE4SZmaWy0nCzMxyOUmYmVkuJwkzM8vlJGFmZrmcJMzMLJeThJmZ5Sp6+dK5ktZL\n2iRpi6QVqfxPJX0pla+VdGrTNcsl7ZC0TdIFRcZnZmaTK3xlOkknRMQ+SbOBzwNXAVsj4vF0/Erg\nnIj4A0nnALcCi4EFwDrgrInL0HllOjMb161LhXajrlyZLiL2pc25wJysKEsQyTOAg2n7YmB1RByI\niBqwA1hSdIxmNjONjq6hXO5n6dIrKJf7GR1d0+mQjjmFJwlJsyRtAvYA90fExlT+bknfAF4HvCOd\nPh/Y1XT57lRmZvY09XqdoaFlNBpj7N37EI3GGENDy6jX650O7Zgyp+gbRMRB4HxJ84A7JJ0TEVsj\n4jrgOklvB64Ehtupd3j4qdMrlQqVSmXKYjaz7ler1ejtXUijMZBKBujpKVOr1dztlFSrVarV6lHV\nUfiYxNNuJl0P/CAi3t9U9gLgUxExIOkasu6oG9OxtcCKiFg/oR6PSZgd5+r1OuVyP43GGDAAPEKp\nNMjOndudJHJ03ZiEpFMknZS2S8BSYLukM5tOuxTYnrbvBC6X1CvpdOBMYEORMZrZzNTX18fIyEpK\npUHmzVtEqTTIyMhKJ4gpVmhLQtK5wCqyZDQLWBMRfybpduBFZAPWO4ErIuLf0jXLgSFgP3B1RNx3\niHrdkjAzwE83teNIWhLT2t00VZwkzMza13XdTWZmNrM5SZiZWS4nCTMzy+UkYWZmuZwkzMwsl5OE\nmZnlcpIwM7NcThJmZpbLScLMzHI5SZiZWS4nCTMzy+UkYWZmuZwkzMwsl5OEmZnlcpIwM7NcThJm\nZpar6OVL50paL2mTpC2SVqTy90raJmmzpH+SNK/pmuWSdqTjFxQZn5mZTa7wlekknRAR+yTNBj4P\nXAXMAx6IiIOSbgAiIpZLOge4FVgMLADWAWdNXIbOK9OZdYaXCp3ZunJluojYlzbnAnOyolgXEQdT\n+YNkCQHgYmB1RByIiBqwA1hSdIxmdnijo2sol/tZuvQKyuV+RkfXdDokmwaHTRKSflPSM9P2dZI+\nIWlRqzeQNEvSJmAPcH9EbJxwyhuBu9P2fGBX07HdqczMOqherzM0tIxGY4y9ex+i0RhjaGgZ9Xq9\n06FZwea0cM71EfFxST8P/Arw58DfAC9v5QapxXB+Gne4Q9I5EbEVQNKfAPsjYrTdwIeHh5/crlQq\nVCqVdqswsxbVajV6exfSaAykkgF6esrUajV3O3WxarVKtVo9qjoOOyYhaVNEnC/pPcCWiPjYeFnb\nN5OuB34QEe+X9Abg94BXRcQT6fg1ZN1RN6b9tcCKiFg/oR6PSZhNo3q9TrncT6MxBgwAj1AqDbJz\n53YniRmkqDGJ3ZI+BLwWuFvS3BavQ9Ipkk5K2yVgKbBd0oXA24CLxxNEcidwuaReSacDZwIbWv9y\nzKwIfX19jIyspFQaZN68RZRKg4yMrHSCOA600pI4AbiQrBWxQ9JpwLkRcd9hK5fOBVaRJZVZwJqI\n+DNJO4Be4Dvp1AcjYlm6ZjkwBOwHrj7UfdySMOsMP900sx1JSyI3SUh69mQXRsR327nRVHKSMDNr\n35EkickGrh8CAjhUhQGc0c6NzMxs5in8ZboiuCVhZta+KW1JHO5diIh4uJ0bmZnZzDPZmMTYJNdF\nRLyqmJAOzy0JM7P2TenAdTdzkjAza18h70lIOiFNx/F3af8sSf/9SIM0M7OZo5WX4j4K/Aj4ubS/\nG3h3YRGZmVnXaCVJvDAi3kv2ctv4rK5tNVfMzGxmaiVJ/ChNqREAkl4IPDH5JWZmdixoZRbYFcBa\n4AWSbgVeCbyhyKDMzKw7tPR0k6TnAK8g62Z6MCL+o+jADhOPn24yM2vTlD7dJKk8PoNrRHwH2Ec2\ni+vrJfUeVaRmZjYjTDYmcRvwDABJ5wEfB74BvBRYWXxoZmbWaZONSZQi4ltp+38At0TETZJmAZuL\nD83MzDptspZEc7/Vq4BPw5PLkZqZ2XFgsiTxgKTbJN0MPAt4ACAtOvSjViqXNFfSekmbJG2RtCKV\nv0bSlyX9eOJEgpKWS9ohaZukC47syzIzs6kw2QR/Iluy9DTgtojYncrPB54bEfe2dAPphIjYJ2k2\n8HngKmAvcBD4EPBH4zPKSjob+BiwGFgArAPOmvgok59uMjNr35ROFZ5+Cq8+RPmmdm6Q3tAGmJvu\nFxHxKDyZiJpdAqyOiANALS1zugRY3849zWY6LxNq3aKVN66PiqRZkjYBe4D7I2LjJKfPB3Y17e9O\nZWbHjdHRNZTL/SxdegXlcj+jo2s6HZIdxwpPEhFxMCLOJ+s+ermkc4q+p9lMVa/XGRpaRqMxxt69\nD9FojDE0tIx6vd7p0Ow4NdnKdJ+OiF+WdGNEvP1obxQRj6WFjC4Etuactht4QdP+glT2E4aHh5/c\nrlQqVCqVow3RrONqtRq9vQtpNAZSyQA9PWVqtZq7naxt1WqVarV6VHVMNnC9FfhfwAjwOibM/NrK\n8qWSTgH2R8TeNEngvcANEXF3Oj5GNnD9UNo/B7gVeDlZN9P9eODajiP1ep1yuZ9GYwwYAB6hVBpk\n587tThJ21KZ04Bp4B3A92W/z759wLMjenTic04BV6QW8WcCaiLhb0qXAB4FTgLskbY6IiyJiq6Tb\nyFoa+4FlzgZ2POnr62NkZCVDQ4P09JTZv38nIyMrnSCsYw47wZ+k6yPiXdMUT0vckrBjnZ9usiIU\ntsa1pIuBX0y71Yi46wjimzJOEmZm7SskSUh6D9m7Cremot8CNkbEtUcU5RRwkjAza19RSeIR4Lzx\nOZvSm9ObImJg0gsL5CRhZta+KV1PYoKTm7ZPaucGZmY2c7WyfOl7gE3pcVWRjU1cU2hUZmbWFVod\nuD6NbNI9gA0RsafQqA4fj7ubzMzaVNjTTd3GScLMrH1FjkmYmdlxyEnCzMxyTZokJM2WtH26gjEz\ns+4yaZKIiB8Dj0r6qWmKx8zMukgrj8A+C/iKpA3AD8YLI+LiwqIyM7Ou0EqSuL7wKMzMrCu1+p5E\nmWxdh3WSTgBmR8T3C48uPx4/Amtm1qZCHoGV9HvA7cCHUtF84I72wzMzs5mmlUdg3wS8EngMICJ2\nAM8tMigzM+sOrSSJJyLiR+M7kuaQrUx3WJLmSlovaZOkLZJWpPJnSbpP0qOS7pV0UtM1yyXtkLRN\n0gXtfkFmZjZ1WkkSn5F0LVCStBT4OPAvrVQeEU8AgxFxPnAecJGkJWQTBK6LiBcDDwDL4ck1ri8D\nzgYuAlZKaqv/zMzMpk4rSeIaoA5sAX4fuBu4rtUbRMS+tDmX7GmqAC4BVqXyVcClaftiYHVEHIiI\nGrCDbMEjs8LU63U2btxIvV7vdChmXeewSSItNrQKeBfwTmBVO48WSZolaROwB7g/IjYCz4uIb6f6\n9/DUGMd8YFfT5btTmVkhRkfXUC73s3TpFZTL/YyOrul0SGZdpZWnm/4b8K/AB4C/Ar4q6aJWbxAR\nB1N30wJgiaSX8JNjGn6e1aZdvV5naGgZjcYYe/c+RKMxxtDQMrcozJq08jLdTWTjCl8FkPRC4FPA\nPe3cKCIek1QFLgS+Lel5EfFtSacC/55O2w28oOmyBansJwwPDz+5XalUqFQq7YRjRq1Wo7d3IY3G\n+Eq8A/T0lKnVavT19XU0NrOpUK1WqVarR1VHK2tcb4yIxU37Ilt4aPEkl42fewqwPyL2SioB9wI3\nAL8EfDcibpT0duBZEXFNGri+FXg5WTfT/WQv8cWEev0ynR21er1OudxPozEGDACPUCoNsnPndicJ\nOyYdyct0uS0JSa9Om1+UdDdwG1m30G8CG1us/zRglaRZZF1bayLibkkPArdJeiOwk+yJJiJiq6Tb\ngK3AfmCZs4EVpa+vj5GRlQwNDdLTU2b//p2MjKx0gjBrktuSkPTRyS6MiN8tJKIWuCVhU6ler1Or\n1Vi4cKEThB3TvHypmZnlmtLupqZKTweuBBY2n++pws3Mjn2tPN10BzBC9pb1wWLDMTOzbtJKkvhh\nRHyg8EjMzKzrtPII7OuAs4D7gCfGyyPi4WJDmzQmj0mYmbWpkDEJ4FzgfwKv4qnupkj7ZmZ2DGul\nJfFV4Jzm6cI7zS0JM7P2FbIyHfBl4OQjC8nMzGayVrqbTga2S9rI08ck/AismdkxrpUksaLwKMzM\nrCv5jWszs+NEUW9cf5+n1nvoBXqAH0TEvPZDNDOzmeSwSSIinjm+naYJvwR4RZFBmZlZdzii7iZJ\nm9Jqcx3h7iYzs/YV1d306qbdWcDLgB+2GZuZmc1ArTzd9GtN2weAGlmXk5mZHeMKfbpJ0gLgH4Dn\nkU3p8eGI+ICklwJ/AzyDLOn8dkQ8nq5ZDryRLCFdHRH3HaJedzeZmbVpShcdkvSOSa6LiHhXCwGd\nCpwaEZslnQh8Efh1YBXwloj4nKQ3AGdExDua1rheDCwA1uE1ro8pXgXOrHOmelqOHxziAzAEvL2V\nyiNiT0RsTtuPA9uB+WQ/+D+XTlsH/EbavhhYHREHIqIG7ACWtPzVWFcbHV1DudzP0qVXUC73Mzq6\nptMhmdlh5CaJiLhp/AP8HVACfhdYDZzR7o0kLQTOAx4EviJpfFqPy8haDZAlkF1Nl+1OZTbD1et1\nhoaW0WiMsXfvQzQaYwwNLaNer3c6NDObxKQD15KeDbwF+G2yLqJFEfGf7d4kdTXdTjbG8LikIeAD\nkq4H7gTanmF2eHj4ye1KpUKlUmm3CptGtVqN3t6FNBoDqWSAnp4ytVrN3U5mBalWq1Sr1aOqY7Ix\niT8HXk3Wivjr8YHltm8gzQHuAu6JiJsPcfws4B8j4hWSriEb77gxHVsLrIiI9ROu8ZjEDFOv1ymX\n+2k0xoAB4BFKpUF27tzuJGE2TaZ6TOKtwPOB64BvSXosfb4v6bE27nELsLU5QUjqS3/OSvX/bTp0\nJ3C5pF5JpwNnAhvauJd1qb6+PkZGVlIqDTJv3iJKpUFGRlY6QZh1uaIfgX0l8FlgC9n8TwFcC7wI\neFPa/0REXNt0zXKywfH9+BHYY46fbjLrnCl9BLabOUmYmbWvqJXpzMzsOOUkYWZmuZwkzMwsl5OE\nmZnlcpIwM7NcThJmZpbLScLMzHI5SZiZWS4nCTMzy+UkYWZmuZwkzMwsl5OEmZnlcpIwM7NcThJm\nZpbLScLMzHI5SZiZWa5Ck4SkBZIekPQVSVskXZXKXyrpC5I2Sdog6WVN1yyXtEPSNkkXFBmfmZlN\nrujlS08FTo2IzZJOBL4I/Drwl8BNEXGfpIuAP46IQUnnALcCi4EFwDrgrInL0HllutZ4qVAza9Z1\nK9NFxJ6I2Jy2Hwe2A88HDgInpdNOBnan7YuB1RFxICJqwA5gSZExHqtGR9dQLvezdOkVlMv9jI6u\n6XRIZjYDTdsa15IWAlXgp8laCfcCSp+fi4hdkj4IfCEiPpau+Qhwd0R8YkJdbklMol6vUy7302iM\nAQPAI5RKg+zcud0tCrPj2JG0JOYUFUyz1NV0O3B1RDwu6Q/S9h2SXgPcAixtp87h4eEntyuVCpVK\nZeoCnuFqtRq9vQtpNAZSyQA9PWVqtZqThNlxpFqtUq1Wj6qOwlsSkuYAdwH3RMTNqex7EXFy0znf\ni4iTJV0DRETcmMrXAisiYv2EOt2SmIRbEmZ2KF03JpHcAmwdTxDJbkm/BCDpl8nGHgDuBC6X1Cvp\ndOBMYMM0xHhM6evrY2RkJaXSIPPmLaJUGmRkZKUThJm1reinm14JfBbYAkT6XAs8BnwAmA38EFgW\nEZvSNcuBIWA/WZfUfYeo1y2JFvjpJjNrdiQtiWkbuJ5KThJmZu3r1u4mMzOboZwkzMwsl5OEmZnl\ncpIwM7NcThJmZpbLScLMzHI5SZiZWS4nCTMzy+UkYWZmuZwkzMwsl5OEmZnlcpIwM7NcThJmZpbL\nScLMzHI5SZiZWS4nCTMzy1VokpC0QNIDkr4iaYukK1P5akkPp8/XJT3cdM1ySTskbZN0QZHxmZnZ\n5OYUXP8B4C0RsVnSicBDku6PiMvHT5D0PuB7afts4DLgbGABsE7SWTNhGTovFWpmx6JCWxIRsSci\nNqftx4FtwPwJp10GfCxtXwKsjogDEVEDdgBLioxxKoyOrqFc7mfp0isol/sZHV3T6ZDMzKbEtI1J\nSFoInAesbyr7BWBPRHwtFc0HdjVdtpufTCpdpV6vMzS0jEZjjL17H6LRGGNoaBn1er3ToZmZHbWi\nu5sASF1NtwNXpxbFuN8CRo+kzuHh4Se3K5UKlUrlKCI8crVajd7ehTQaA6lkgJ6eMrVazd1OZtZR\n1WqVarV6VHWo6O5+SXOAu4B7IuLmpvLZZC2FRRHxrVR2DRARcWPaXwusiIj1E+rsmmGKer1OudxP\nozEGDACPUCoNsnPndicJM+sqkogItXPNdHQ33QJsbU4QyVJg23iCSO4ELpfUK+l04ExgwzTEeMT6\n+voYGVlJqTTIvHmLKJUGGRlZ6QRhZseEQlsSkl4JfBbYAkT6XBsRayV9FPhCRPzdhGuWA0PAfrLu\nqfsOUW/XtCTG+ekmM+t2R9KSKLy7qQjdmCTMzLpdt3Y3mZnZDOUkYWZmuZwkzMwsl5OEmZnlcpIw\nM7NcThJmZpbLScLMzHI5SZiZWS4nCTMzy+UkYWZmuZwkzMwsl5OEmZnlcpIwM7NcThJmZpbLScLM\nzHIVmiQkLZD0gKSvSNoi6aqmY1dK2pbKb2gqXy5pRzp2QZHxmZnZ5IpuSRwA3hIRLwF+FniTpH5J\nFeDXgHMj4lzgfQCSzgYuA84GLgJWSmprgYxOOtoFx4vgmFrjmFrXjXE5puIUmiQiYk9EbE7bjwPb\ngPnAHwA3RMSBdOw/0iWXAKsj4kBE1IAdwJIiY5xK3fhN4Zha45ha141xOabiTNuYhKSFwHnAeuBF\nwC9KelDSmKSfSafNB3Y1XbY7lZmZWQfMmY6bSDoRuB24OiIelzQHeFZEvELSYuDjwBnTEYuZmbVO\nEVHsDbKEcBdwT0TcnMruBm6MiM+k/R3AK4DfA4iIG1L5WmBFRKyfUGexQZuZHaMioq1x3uloSdwC\nbB1PEMkdwKuAz0h6EdAbEd+RdCdwq6T3k3UznQlsmFhhu1+kmZkdmUKThKRXAr8NbJG0CQjgWuCj\nwC2StgBPAK8HiIitkm4DtgL7gWVRdFPHzMxyFd7dZGZmM1fXv3F9iBfyrkzlKyR9U9LD6XNhB2M6\n7EuCHYhp/O9pddPf0dclPdzBmK5K5S+V9AVJmyRtkPSy6Yqphbj+r6QvSfpkeuBiumKaK2l9+jvZ\nImlFKn+WpPskPSrpXkkndUFMr5H0ZUk/lrRouuI5TEzvTf/vNkv6J0nzuiCmP03fS5skrZV0aqdj\najr+VkkHJT37sJVFRFd/gFOB89L2icCjQD+wguxFvW6KqQLcB8xJx07pdEwTznkfcF0HY9pO9qLk\nvcAFqfwiYKzD/37jcW0Afj6VvwH402mO64T052zgQbJ3hG4E/jiVv53s/aJOx/Ri4CzgAWDRdMYz\nSUy/AsxK5TcA7+mCmE5sOn4l8DedjintLwDWAl8Hnn24erq+JRH5L+QBdGQAe5KY8l4S7GRMzS4D\nRjsY03bg+cBBYPw34pPJ3oeZNjlxzQfOiojPpdPWAb8xzXHtS5tzycYLg+wF01WpfBVwaadjiohH\nI2IHnfv/d6iY1kXEwVT+INkPwk7H9HjTKc8g+77vaExp/y+At7VaT9cniWYTXsgDeHNqXn5kOpvh\nk8Q08SXBae1GyYlpvOwXgD0R8a9dENMfAu+T9A3gvcDyTsQ0Ia4Hga9Iujgduoxp/kEjaVZ6wGMP\ncH9EbASeFxHfhiy5Ac/tgpg6qoWY3gjc0w0xSXp3+j5/HfCOTseUvr93RcSWliuazubPUTadTgS+\nCFyS9vt4auD93cBIF8S0Bbg5bS8GvtbpmJrKVwJ/2CX/djcDl6bt16Rv4G6I68VkXWEbgeuBeofi\nmgd8GngJ8N0Jx77TwZgeAM5pKhujA91Nh4npT4B/6qaYUvnbgeEOfz+dS/bL0DNT+deB5xzu+hnR\nklD2Qt7twD9GxCcBIqIe6SsFPkz2Q7mjMZFNKfKJFN9G4KCk53Q4JiTNBl4NrJmuWA4T0+9ExB0A\nEXE7HZifK+d76tGI+NWIWAysBjrS6oqIx4AqcCHwbUnPSzGfCvx7B2MaSzF1hYkxSXoD8F/Jfmvv\nipiafIxp7r4c1/T9dAmwEPiSpK+TtZQfkjRp63RGJAkO8ULehCcFXg18udMx8dRLgih7SbAnIr7T\n4ZgAlgLbIuJb0xjLZDHtlvRLAJJ+Gfh/3RCXpL705yzgOuBvpysYSaeMd5lKKpH+zYA7yQbRAX4H\n+OQhK5i+mLZPPG264pksJmVPN74NuDginuiSmM5sOu1Ssn/PTsb0cEScGhFnRMTpwDeB8yNi0l88\npmXupqOh/BfyXifpPLLBoBrw+10Q0yFfEuxkTBGxFngt0zhgfbiYyKZf+UBq4fwQ+N9dEteLJL0p\n7X8iIv5+GsM6DViVEtQsYE1E3C3pQeA2SW8EdpKNlXQ6pkuBDwKnAHdJ2hwRF3U4ph1AL3C/stUF\nHoyIZR2O6fb0y+JBsn+7K6YpntyYJpwTtJDk/TKdmZnlmindTWZm1gFOEmZmlstJwszMcjlJmJlZ\nLicJMzPL5SRhZma5uv49CbN2pKmPP032DPhpwI+BetpfEmnyxW4i6XeBTx3upSazTvB7EnbMkvQO\n4PGIeH8XxDIrnpqldOKx/wO8OSK+1EZ9syPix1MWoFkOdzfZsexpb5NKen1aiOVhSX+VymZL+k9J\nNylbSOceSUskVSV9NU33gKQhSZ9I5Y9K+pMW6/0LSZuBxZKGlS2y9Iiklem8y8hmoR1fHKpH0i6l\nRXMkvVzS/Wn7XZJWSfoc8NF0j5uUzTq8Ob2VbTalnCTsuCDpJcCvAz8bEYuAHkmXp8MnkXX3/DTZ\n2uoryObgugx4V1M1i4GLgfPJpoUZaKHeakScFxHrgb+MiCURMQCcLOlXI+I2YDNwWUQsioj9PDXv\n/7jm/RcDgxHxerLpTL4dEa8gmyTxzZKmdXpzO/Z5TMKOF78CvAz4orLJff4L2Xw6APsi4oG0vQX4\nXkQcTHNwlZvquDfNqImkfwZ+HuiZpN4nmmfjBZZK+qN0znPIpim/Nx1rbvVMNp/OJ1MiAbgA6Jf0\nW2l/HtmKcd+c5HqztjhJ2PFCwC0RMXGt39nAj5qKDpJNzji+3fx/pPk3ejXt59XbaNovkU2Kd15E\n7JH0LrJkcSgHeKqVP/GcH0yIYVlEjOXUY3bU3N1kx4t1wGVK63tIenZT18xkv7k3H7tA0jxJJ5DN\nzf95siepmADDAAAA0UlEQVSpWqm3RPak1XckPZOnry3wfbJWwLivAz+Ttidbg+Be4E0pISHpRZLm\nTnK+WdvckrDjQkR8WdI7gXVp+uQfkU3d/G/85BjA0y5t2t5Itr7DacDfR8QjAK3UGxHflbSKbE2B\nb5GtEDbuo8BHJO0jG1t4J/BhSf8JfHaS2D4E/BSwWVKQLUh0CU+1hMyOmh+BNWuBpCHgJRHxlk7H\nYjad3N1kZma53JIwM7NcbkmYmVkuJwkzM8vlJGFmZrmcJMzMLJeThJmZ5XKSMDOzXP8ftdld3yZF\nH54AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1167cb110>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.plot(kind=\"scatter\", x=\"Temperature\", y=\"Number of Sells\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4) scipy 모듈을 활용한 선형 회귀 분석"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "slope, intercept, r_value, p_value, std_err = stats.linregress(df['Temperature'].values, df['Number of Sells'].values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                   slope: 10.0000000000\n",
      "                               intercept: 10.0000000000\n",
      "       r_value (Correlation Coefficient): 1.0000000000\n",
      "r-squared (Coefficient of Determination): 1.0000000000\n",
      "           p_value (Hyperthesis Testing): 0.0000000000\n",
      "                std_err (Standard Error): 0.0000000000\n"
     ]
    }
   ],
   "source": [
    "format = \"%40s: %12.10f\"\n",
    "print format % (\"slope\", slope)\n",
    "print format % (\"intercept\", intercept)\n",
    "print format % (\"r_value (Correlation Coefficient)\", r_value)\n",
    "print format % (\"r-squared (Coefficient of Determination)\", r_value**2)\n",
    "print format % (\"p_value (Hyperthesis Testing)\", p_value)\n",
    "print format % (\"std_err (Standard Error)\", std_err)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 5) 선형 회귀식 및 예측"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 회귀식: $y = intercept + slope \\times x$\n",
    "  - 위 예제의 회귀식: $y = 10.0 + 10.0 \\times x$\n",
    "\n",
    "- 질문: 온도가 34일 때 예상 에어콘 판매량은? $10 + 10 \\times 34 = 350$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 6) 상관 계수 (correlation coefficient)\n",
    "- r-value (Pearson correlation coefficient): \n",
    "  $$ r-value = \\frac{cov(x_i, y_i)}{\\sigma_{x_i} \\sigma_{y_i}}$$\n",
    "  \n",
    "  where \n",
    "  - $cov(x_i,y_i)$ is the covariance of $x_i$ and $y_i$\n",
    "    - corvariance: https://goo.gl/ryS1dO\n",
    "  - $\\sigma_{x_i}$ and $\\sigma_{y_i}$ are the standard deviation of $x_i$ and $y_i$, respectively.\n",
    "- r-value의 의미\n",
    "  - $x_i$ 값들과 $y_i$ 값들 사이의 관계\n",
    "    - 1: 강한 양의 상관 관계\n",
    "    - 0: 아무 관계 없음\n",
    "    - -1: 강한 음의 상관 관계\n",
    "  - 참고: https://en.wikipedia.org/wiki/Pearson_product-moment_correlation_coefficient"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- [참고] Covariance 구하는 Python 코딩"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def cov(a, b):\n",
    "\n",
    "    if len(a) != len(b):\n",
    "        return\n",
    "\n",
    "    a_mean = np.mean(a)\n",
    "    b_mean = np.mean(b)\n",
    "\n",
    "    sum = 0\n",
    "\n",
    "    for i in range(0, len(a)):\n",
    "        sum += ((a[i] - a_mean) * (b[i] - b_mean))\n",
    "\n",
    "    return sum/(len(a) - 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 위 코드에서 마지막 라인에 1을 빼는 이유\n",
    "  - 자유도 (Degree of Feeedom): http://math7.tistory.com/113"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- numpy를 활용한 r-value (상관 계수) 구하기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "60.0\n",
      "2.44948974278\n",
      "24.4948974278\n",
      "1.0\n"
     ]
    }
   ],
   "source": [
    "a = np.cov(df['Temperature'].values, df['Number of Sells'].values, ddof = 1)[0][1]\n",
    "print a\n",
    "b = np.std(df['Temperature'].values, ddof = 1)\n",
    "print b\n",
    "c = np.std(df['Number of Sells'].values, ddof = 1)\n",
    "print c\n",
    "print a / (b * c)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- numpy를 이용한 r-value (상관 계수) 구하기"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.corrcoef(df['Temperature'].values, df['Number of Sells'].values, ddof = 1)[0][1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 7) 결정 계수 (Coefficient of Determination)\n",
    "- r-squared: 추정된 회귀선이 표본 자료에 적합한가를 측정하는 결정 계수\n",
    "- 0 에서 1 사이에 값을 갖는다\n",
    "  - r-squared이 0에 가까울 때: 두 변수 X와 Y 사이에는 관계가 거의 없으며 회귀분석을 적용함에 적합치 않다는 의미\n",
    "  - r-squared이 1에 가까울 때: 두 변수 X와 Y 사이에는 매우 강한 관계가 있으며 회귀분석 적용이 적합함을 의미\n",
    "    - r-squared가 1이라는 것은 두 변수 X와 Y가 확정적(Deterministic) 관계이며 그 관계성에 오차가 없으므로 회귀분석을 적용할 이유가 없다."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 8) 가설 검증 (Hyperthesis Testing)과 p-value (유의 확률)\n",
    "- 다음의 두 가지 가설을 세울 수 있음\n",
    "  - **Null Hyperthesis** (귀무 가설): 예측 변수(Predictor)와 결과 변수(Outcome Variable)사이에 아무런 관계가 없다.\n",
    "  - **Alternative Hypothesis** (대립 가설): 예측 변수(Predictor)와 결과 변수(Outcome Variable)사이에 (선형)관계가 존재한다.\n",
    "- p-value (유의 확률): **Null Hyperthesis**가 만족할 확률\n",
    "- 통계학적으로, **p-value < 0.05** 이면 **Null Hyperthesis**가 만족할 확률은 거의 없다고 간주하고 **Alternative Hyperthesis**를 받아들인다.\n",
    "  - 참고: http://www.editage.co.kr/edm-nw-v6.html\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 9) 표준 오차 (Standard Error)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 표준 오차 (std_err): 통계치 ($x_i$, $y_i$)들과 선형 회귀 모델사이의 평균 거리\n",
    "- 표준 오차가 적을 수록 그러한 통계치가 선형 회귀 모델식에 가깝게 분포"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Simple Linear Regression - II (실제 통계 데이터를 활용한 선형 회귀분석)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1) 데이터 설명\n",
    "- URL\n",
    "  - https://raw.githubusercontent.com/bluebibi/LINK_ML_BIG_DATA/master/death_rate.csv\n",
    "- D: 사망률\n",
    "- A1 ~ A15: 사망률에 영향이 있을 것 같은 각종 요인 데이터\n",
    "- References\n",
    "  - Richard Gunst and Robert Mason, Regression Analysis and Its Applications: a data-oriented approach, Dekker, 1980, pages 370-371\n",
    "  - Gary McDonald and Richard Schwing, Instabilities of regression estimates relating air pollution to mortality, Technometrics, Volume 15, Number 3, pages 463-482, 1973.\n",
    "  - Helmut Spaeth, Mathematical Algorithms for Linear Regression, Academic Press, 1991, ISBN 0-12-656460-4.\n",
    "    \n",
    "- The death rate is to be represented as a function of other variables.\n",
    "- There are 60 rows of data. The data includes:\n",
    "  - I: the index;\n",
    "  - A1: the average annual precipitation (강수량);\n",
    "  - A2: the average January temperature;\n",
    "  - A3: the average July temperature;\n",
    "  - A4: the size of the population older than 65;\n",
    "  - A5: the number of members per household;\n",
    "  - A6: the number of years of schooling for persons over 22;\n",
    "  - A7: the number of households with fully equipped kitchens;\n",
    "  - A8: the population per square mile; \n",
    "  - A9: the size of the nonwhite population;\n",
    "  - A10: the number of office workers;\n",
    "  - A11: the number of families with an income less than $3000;\n",
    "  - A12: the hydrocarbon pollution index;\n",
    "  - A13: the nitric oxide pollution index;\n",
    "  - A14: the sulfur dioxide pollution index;\n",
    "  - A15: the degree of atmospheric moisture.\n",
    "  - D: the death rate."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2) Pandas를 활용한 데이터프레임 정의"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "import urllib2\n",
    "import json\n",
    "\n",
    "path = 'https://raw.githubusercontent.com/bluebibi/LINK_ML_BIG_DATA/master/death_rate.csv'\n",
    "raw_csv = urllib2.urlopen(path)\n",
    "df = pd.read_csv(raw_csv)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>I</th>\n",
       "      <th>A1</th>\n",
       "      <th>A2</th>\n",
       "      <th>A3</th>\n",
       "      <th>A4</th>\n",
       "      <th>A5</th>\n",
       "      <th>A6</th>\n",
       "      <th>A7</th>\n",
       "      <th>A8</th>\n",
       "      <th>A9</th>\n",
       "      <th>A10</th>\n",
       "      <th>A11</th>\n",
       "      <th>A12</th>\n",
       "      <th>A13</th>\n",
       "      <th>A14</th>\n",
       "      <th>A15</th>\n",
       "      <th>D</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>36</td>\n",
       "      <td>27</td>\n",
       "      <td>71</td>\n",
       "      <td>8.1</td>\n",
       "      <td>3.34</td>\n",
       "      <td>11.4</td>\n",
       "      <td>81.5</td>\n",
       "      <td>3243</td>\n",
       "      <td>8.8</td>\n",
       "      <td>42.6</td>\n",
       "      <td>11.7</td>\n",
       "      <td>21</td>\n",
       "      <td>15</td>\n",
       "      <td>59</td>\n",
       "      <td>59</td>\n",
       "      <td>0.921</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>35</td>\n",
       "      <td>23</td>\n",
       "      <td>72</td>\n",
       "      <td>11.1</td>\n",
       "      <td>3.14</td>\n",
       "      <td>11.0</td>\n",
       "      <td>78.8</td>\n",
       "      <td>4281</td>\n",
       "      <td>3.6</td>\n",
       "      <td>50.7</td>\n",
       "      <td>14.4</td>\n",
       "      <td>8</td>\n",
       "      <td>10</td>\n",
       "      <td>39</td>\n",
       "      <td>57</td>\n",
       "      <td>0.997</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>44</td>\n",
       "      <td>29</td>\n",
       "      <td>74</td>\n",
       "      <td>10.4</td>\n",
       "      <td>3.21</td>\n",
       "      <td>9.8</td>\n",
       "      <td>81.6</td>\n",
       "      <td>4260</td>\n",
       "      <td>0.8</td>\n",
       "      <td>39.4</td>\n",
       "      <td>12.4</td>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "      <td>33</td>\n",
       "      <td>54</td>\n",
       "      <td>0.962</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>47</td>\n",
       "      <td>45</td>\n",
       "      <td>79</td>\n",
       "      <td>6.5</td>\n",
       "      <td>3.41</td>\n",
       "      <td>11.1</td>\n",
       "      <td>77.5</td>\n",
       "      <td>3125</td>\n",
       "      <td>27.1</td>\n",
       "      <td>50.2</td>\n",
       "      <td>20.6</td>\n",
       "      <td>18</td>\n",
       "      <td>8</td>\n",
       "      <td>24</td>\n",
       "      <td>56</td>\n",
       "      <td>0.982</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>43</td>\n",
       "      <td>35</td>\n",
       "      <td>77</td>\n",
       "      <td>7.6</td>\n",
       "      <td>3.44</td>\n",
       "      <td>9.6</td>\n",
       "      <td>84.6</td>\n",
       "      <td>6441</td>\n",
       "      <td>24.4</td>\n",
       "      <td>43.7</td>\n",
       "      <td>14.3</td>\n",
       "      <td>43</td>\n",
       "      <td>38</td>\n",
       "      <td>206</td>\n",
       "      <td>55</td>\n",
       "      <td>1.071</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   I  A1  A2  A3    A4    A5    A6    A7    A8    A9   A10   A11  A12  A13  \\\n",
       "0  1  36  27  71   8.1  3.34  11.4  81.5  3243   8.8  42.6  11.7   21   15   \n",
       "1  2  35  23  72  11.1  3.14  11.0  78.8  4281   3.6  50.7  14.4    8   10   \n",
       "2  3  44  29  74  10.4  3.21   9.8  81.6  4260   0.8  39.4  12.4    6    6   \n",
       "3  4  47  45  79   6.5  3.41  11.1  77.5  3125  27.1  50.2  20.6   18    8   \n",
       "4  5  43  35  77   7.6  3.44   9.6  84.6  6441  24.4  43.7  14.3   43   38   \n",
       "\n",
       "   A14  A15      D  \n",
       "0   59   59  0.921  \n",
       "1   39   57  0.997  \n",
       "2   33   54  0.962  \n",
       "3   24   56  0.982  \n",
       "4  206   55  1.071  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3) 회귀 분석을 위한 몇가지 요인 선택"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[36 35 44 47 43 53 43 45 36 36 52 33 40 35 37 35 36 15 31 30 31 31 42 43 46\n",
      " 39 35 43 11 30 50 60 30 25 45 46 54 42 42 36 37 42 41 44 32 34 10 18 13 35\n",
      " 45 38 31 40 41 28 45 45 42 38]\n",
      "[ 0.921  0.997  0.962  0.982  1.071  1.03   0.934  0.899  1.001  0.912\n",
      "  1.017  1.024  0.97   0.985  0.958  0.86   0.936  0.871  0.959  0.941\n",
      "  0.891  0.871  0.971  0.887  0.952  0.968  0.919  0.844  0.861  0.989\n",
      "  1.006  0.861  0.929  0.857  0.961  0.923  1.113  0.994  1.015  0.991\n",
      "  0.893  0.938  0.946  1.025  0.874  0.953  0.839  0.911  0.79   0.899\n",
      "  0.904  0.95   0.972  0.912  0.967  0.823  1.003  0.895  0.911  0.954]\n"
     ]
    }
   ],
   "source": [
    "print df['A1'].values\n",
    "print df['D'].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{1: 0.51058981544585491, 2: -0.058248264329082554, 3: 0.279026441842534, 4: -0.17474932763602924, 5: 0.3586407523344754, 6: -0.51252452972931994, 7: -0.42771304952186134, 8: 0.26418567700899515, 9: 0.64413803239793543, 10: -0.28529506633050106, 11: 0.41125099134506982, 12: -0.17826806806438281, 13: -0.081368918173860549, 14: 0.42579785580212309, 15: -0.05410071395571768}\n",
      "\n",
      "[(9, 0.64413803239793543), (1, 0.51058981544585491), (14, 0.42579785580212309), (11, 0.41125099134506982), (5, 0.3586407523344754), (3, 0.279026441842534), (8, 0.26418567700899515), (15, -0.05410071395571768), (2, -0.058248264329082554), (13, -0.081368918173860549), (4, -0.17474932763602924), (12, -0.17826806806438281), (10, -0.28529506633050106), (7, -0.42771304952186134), (6, -0.51252452972931994)]\n"
     ]
    }
   ],
   "source": [
    "corr_dic = {}\n",
    "for i in range(1,16):\n",
    "    corr_dic[i] = np.corrcoef(df['A' + str(i)].values, df['D'].values, ddof = 1)[0][1]\n",
    "print corr_dic\n",
    "\n",
    "print \n",
    "\n",
    "sorted_corr_dic = sorted(corr_dic.items(), key=lambda x: x[1], reverse=True)\n",
    "print sorted_corr_dic"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 선택한 요인\n",
    "  - A9: the size of the nonwhite population\n",
    "  - A1: the average annual precipitation (강수량)\n",
    "  - A6: the number of years of schooling for persons over 22"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df_sub = df[['A9','A1','A6', 'D']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>A9</th>\n",
       "      <th>A1</th>\n",
       "      <th>A6</th>\n",
       "      <th>D</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8.8</td>\n",
       "      <td>36</td>\n",
       "      <td>11.4</td>\n",
       "      <td>0.921</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3.6</td>\n",
       "      <td>35</td>\n",
       "      <td>11.0</td>\n",
       "      <td>0.997</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.8</td>\n",
       "      <td>44</td>\n",
       "      <td>9.8</td>\n",
       "      <td>0.962</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>27.1</td>\n",
       "      <td>47</td>\n",
       "      <td>11.1</td>\n",
       "      <td>0.982</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>24.4</td>\n",
       "      <td>43</td>\n",
       "      <td>9.6</td>\n",
       "      <td>1.071</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     A9  A1    A6      D\n",
       "0   8.8  36  11.4  0.921\n",
       "1   3.6  35  11.0  0.997\n",
       "2   0.8  44   9.8  0.962\n",
       "3  27.1  47  11.1  0.982\n",
       "4  24.4  43   9.6  1.071"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_sub.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 4) 통계 수치에 대한 Scatter Plot 생성"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x119f47950>"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAABBIAAAF6CAYAAACkx3tWAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x119e7a0d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(17, 6))\n",
    "ax1 = fig.add_subplot(131)\n",
    "ax1.scatter(df_sub['A9'], df_sub['D'])\n",
    "ax1.set_title(\"Data Rate vs. Size of the nonwhite population\")\n",
    "\n",
    "ax2 = fig.add_subplot(132)\n",
    "ax2.scatter(df_sub['A1'], df_sub['D'])\n",
    "ax2.set_title(\"Data Rate vs. Average annual precipitation\")\n",
    "\n",
    "ax3 = fig.add_subplot(133)\n",
    "ax3.scatter(df_sub['A6'], df_sub['D'])\n",
    "ax3.set_title(\"Data Rate vs. Number of years of schooling for persons over 22\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 5) scipy 모듈을 활용한 선형 회귀 분석"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "slope, intercept, r_value, p_value, std_err = stats.linregress(df_sub['A9'].values, df_sub['D'].values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                                   slope: 0.0044945782\n",
      "                               intercept: 0.8865010410\n",
      "       r_value (Correlation Coefficient): 0.6441380324\n",
      "r-squared (Coefficient of Determination): 0.4149138048\n",
      "           p_value (Hyperthesis Testing): 0.0000000281\n",
      "                std_err (Standard Error): 0.0007008191\n"
     ]
    }
   ],
   "source": [
    "format = \"%40s: %12.10f\"\n",
    "print format % (\"slope\", slope)\n",
    "print format % (\"intercept\", intercept)\n",
    "print format % (\"r_value (Correlation Coefficient)\", r_value)\n",
    "print format % (\"r-squared (Coefficient of Determination)\", r_value**2)\n",
    "print format % (\"p_value (Hyperthesis Testing)\", p_value)\n",
    "print format % (\"std_err (Standard Error)\", std_err)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 모든 Predicator 변수들에 대한 분석"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "No.           slope       intercept         r_value       r_squared         p_value         std_err\n",
      "  1        0.003183        0.820937        0.510590        0.260702        0.000031        0.000704\n",
      "  2       -0.000303        0.950407       -0.058248        0.003393        0.658436        0.000681\n",
      "  3        0.003644        0.668059        0.279026        0.077856        0.030855        0.001647\n",
      "  4       -0.007426        1.005207       -0.174749        0.030537        0.181736        0.005494\n",
      "  5        0.165038        0.401320        0.358641        0.128623        0.004895        0.056404\n",
      "  6       -0.037738        1.353973       -0.512525        0.262681        0.000028        0.008302\n",
      "  7       -0.005178        1.358817       -0.427713        0.182938        0.000653        0.001437\n",
      "  8        0.000011        0.896037        0.264186        0.069794        0.041378        0.000005\n",
      "  9        0.004495        0.886501        0.644138        0.414914        0.000000        0.000701\n",
      " 10       -0.003838        1.116685       -0.285295        0.081393        0.027137        0.001693\n",
      " 11        0.006153        0.851430        0.411251        0.169127        0.001098        0.001791\n",
      " 12       -0.000121        0.944433       -0.178268        0.031780        0.172960        0.000087\n",
      " 13       -0.000109        0.942326       -0.081369        0.006621        0.536543        0.000176\n",
      " 14        0.000418        0.917388        0.425798        0.181304        0.000694        0.000117\n",
      " 15       -0.000617        0.975347       -0.054101        0.002927        0.681404        0.001495\n"
     ]
    }
   ],
   "source": [
    "predicator_analysis = {}\n",
    "for i in range(1, 16):\n",
    "    predicator_analysis[i] = Series(np.empty(6), index=['slope', 'intercept', 'r_value', 'r_squared', 'p_value', 'std_err'])\n",
    "    predicator_analysis[i][0],\\\n",
    "    predicator_analysis[i][1],\\\n",
    "    predicator_analysis[i][2],\\\n",
    "    predicator_analysis[i][4],\\\n",
    "    predicator_analysis[i][5] = stats.linregress(df['A' + str(i)].values, df['D'].values)\n",
    "    predicator_analysis[i][3] = predicator_analysis[i][2] ** 2\n",
    "\n",
    "\n",
    "format1 = \"%3s %15s %15s %15s %15s %15s %15s\"\n",
    "format2 = \"%3d %15f %15f %15f %15f %15f %15f\"\n",
    "print format1 % ('No.', 'slope', 'intercept', 'r_value', 'r_squared', 'p_value', 'std_err')\n",
    "for i in range(1, 16):\n",
    "    lst = [i]\n",
    "    for j in range(6):\n",
    "        lst.append(predicator_analysis[i][j])\n",
    "    print format2 % tuple(lst)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 6) 선형 회귀식 및 예측"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " - 연간 강수량 (A1)과 사망률간의 관계: $y = 0.82 + 0.003 * x$ \n",
    "   - 강수량이 70일 때 사망률은? $0.82 + 0.003 * 70 = 1.03$"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 7) 선형 회귀식과 Scatter Plot을 함께 생성"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x11b0be310>]"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA+gAAAFwCAYAAAArCVHjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nUgGcoEuSJEmSVID/Ax9v3xV+AWu2AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11a5228d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(17, 6))\n",
    "ax1 = fig.add_subplot(131)\n",
    "ax1.scatter(df_sub['A9'], df_sub['D'])\n",
    "line_plot_x1 = np.linspace(df_sub['A9'].min(), df_sub['A9'].max(), 10)\n",
    "slope, intercept, r_value, p_value, std_err = stats.linregress(df_sub['A9'].values, df_sub['D'].values)\n",
    "ax1.plot(line_plot_x1, intercept + slope * line_plot_x1)\n",
    "\n",
    "ax2 = fig.add_subplot(132)\n",
    "ax2.scatter(df_sub['A1'], df_sub['D'])\n",
    "line_plot_x2 = np.linspace(df_sub['A1'].min(), df_sub['A1'].max(), 10)\n",
    "slope, intercept, r_value, p_value, std_err = stats.linregress(df_sub['A1'].values, df_sub['D'].values)\n",
    "ax2.plot(line_plot_x2, intercept + slope * line_plot_x2)\n",
    "\n",
    "ax3 = fig.add_subplot(133)\n",
    "ax3.scatter(df_sub['A6'], df_sub['D'])\n",
    "line_plot_x3 = np.linspace(df_sub['A6'].min(), df_sub['A6'].max(), 10)\n",
    "slope, intercept, r_value, p_value, std_err = stats.linregress(df_sub['A6'].values, df_sub['D'].values)\n",
    "ax3.plot(line_plot_x3, intercept + slope * line_plot_x3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## 3. Multivariate Linear Regression (다변수 선형 회귀분석) \n",
    "### - sklearn 모듈을 활용한 다중 변수 선형 회귀"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "from sklearn import linear_model\n",
    "regr = linear_model.LinearRegression()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1) A9과 A1에 대한 사망률 변화\n",
    " - A9: the size of the nonwhite population\n",
    " - A1: the average annual precipitation (강수량)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>A9</th>\n",
       "      <th>A1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>8.8</td>\n",
       "      <td>36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3.6</td>\n",
       "      <td>35</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.8</td>\n",
       "      <td>44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>27.1</td>\n",
       "      <td>47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>24.4</td>\n",
       "      <td>43</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     A9  A1\n",
       "0   8.8  36\n",
       "1   3.6  35\n",
       "2   0.8  44\n",
       "3  27.1  47\n",
       "4  24.4  43"
      ]
     },
     "execution_count": 130,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[['A9', 'A1']].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[(8.8000000000000007, 36), (3.6000000000000001, 35), (0.80000000000000004, 44), (27.100000000000001, 47), (24.399999999999999, 43), (38.5, 53), (3.5, 43), (5.2999999999999998, 45), (8.0999999999999996, 36), (6.7000000000000002, 36), (22.199999999999999, 52), (16.300000000000001, 33), (13.0, 40), (14.699999999999999, 35), (13.1, 37), (14.800000000000001, 35), (12.4, 36), (4.7000000000000002, 15), (15.800000000000001, 31), (13.1, 30), (11.5, 31), (5.0999999999999996, 31), (22.699999999999999, 42), (7.2000000000000002, 43), (21.0, 46), (15.6, 39), (12.6, 35), (2.8999999999999999, 43), (7.7999999999999998, 11), (13.1, 30), (36.700000000000003, 50), (13.6, 60), (5.7999999999999998, 30), (2.0, 25), (21.0, 45), (8.8000000000000007, 46), (31.399999999999999, 54), (11.300000000000001, 42), (17.5, 42), (8.0999999999999996, 36), (3.6000000000000001, 37), (2.2000000000000002, 42), (2.7000000000000002, 41), (28.600000000000001, 44), (5.0, 32), (17.199999999999999, 34), (5.9000000000000004, 10), (13.699999999999999, 18), (3.0, 13), (5.7000000000000002, 35), (3.3999999999999999, 45), (3.7999999999999998, 38), (9.5, 31), (2.5, 40), (25.899999999999999, 41), (7.5, 28), (12.1, 45), (1.0, 45), (4.7999999999999998, 42), (11.699999999999999, 38)]\n"
     ]
    }
   ],
   "source": [
    "X = zip(df['A9'], df['A1'])\n",
    "print X\n",
    "y = df['D'].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Coefficients: [ 0.00364445  0.00183603]\n",
      "Intercept: 0.827988572515\n"
     ]
    }
   ],
   "source": [
    "regr = regr.fit(X, y)\n",
    "\n",
    "print 'Coefficients:', regr.coef_\n",
    "print 'Intercept:', regr.intercept_"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2) 선형 회귀식 및 예측"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 선형 회귀식: $y = 0.8280 + 0.0036 * A9 + 0.0018 * A1$ "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.98122101]\n",
      "0.9792\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/yhhan/anaconda/lib/python2.7/site-packages/sklearn/utils/validation.py:386: DeprecationWarning: Passing 1d arrays as data is deprecated in 0.17 and willraise ValueError in 0.19. Reshape your data either using X.reshape(-1, 1) if your data has a single feature or X.reshape(1, -1) if it contains a single sample.\n",
      "  DeprecationWarning)\n"
     ]
    }
   ],
   "source": [
    "test_x = [36, 12]\n",
    "print regr.predict(test_x)\n",
    "print 0.8280 + 0.0036 * test_x[0] + 0.0018 * test_x[1] "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3) Scatter Plot 및 선형 회귀 Plane 생성"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<mpl_toolkits.mplot3d.art3d.Poly3DCollection at 0x124c9e510>"
      ]
     },
     "execution_count": 134,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA8IAAAGUCAYAAAD+o5tuAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x125358e90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot outputs\n",
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "fig = plt.figure(figsize=(17, 7))\n",
    "ax1 = fig.add_subplot(121, projection='3d')\n",
    "ax1.scatter(df['A9'], df['A1'], df_sub['D'])\n",
    "\n",
    "ax2 = fig.add_subplot(122, projection='3d')\n",
    "ax2.scatter(df['A9'], df['A1'], df_sub['D'])\n",
    "# create x,y\n",
    "xx, yy = np.meshgrid(range(int(df['A9'].min()), int(df['A9'].max())), range(int(df['A1'].min()), int(df['A1'].max())))\n",
    "# calculate corresponding z\n",
    "z = 0.8280 + 0.0036 * xx + 0.0018 * yy\n",
    "ax2.plot_surface(xx, yy, z, rstride=1, cstride=1, linewidth=0, color=\"yellow\", shade=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 4. Tensorflow를 활용한 회귀분석"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<type 'numpy.ndarray'> (2, 100)\n",
      "<type 'numpy.ndarray'> (100,)\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "import numpy as np\n",
    "\n",
    "# Numpy 랜덤으로 100개의 가짜 데이터 채우기.\n",
    "x_data = np.float32(np.random.rand(2, 100))\n",
    "# 학습 레이블(목표값)은 아래의 식으로 산출. (W = [0.1, 0.2], b = 0.3)\n",
    "y_data = np.dot([0.100, 0.200], x_data) + 0.300\n",
    "print type(x_data), x_data.shape\n",
    "print type(y_data), y_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<type 'numpy.ndarray'> (2, 60)\n",
      "<type 'numpy.ndarray'> (60,)\n"
     ]
    }
   ],
   "source": [
    "import tensorflow as tf\n",
    "import numpy as np\n",
    "\n",
    "x_data = df[['A9', 'A1']].T\n",
    "y_data = df['D']\n",
    "x_data = x_data.as_matrix().astype('float32')\n",
    "y_data = y_data.as_matrix().astype('float32')\n",
    "print type(x_data), x_data.shape\n",
    "print type(y_data), y_data.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 입력 데이터와 W, b를 사용해 선형 모델 정의"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(1, 2)\n",
      "(1, 60)\n"
     ]
    }
   ],
   "source": [
    "# b는 0,\n",
    "b = tf.Variable(tf.zeros([1]))\n",
    "# W는 1x2 형태의 웨이트 변수\n",
    "W = tf.Variable(tf.zeros([1, 2]))\n",
    "y = tf.matmul(W, x_data) + b\n",
    "print W.get_shape()\n",
    "print y.get_shape()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 손실과 학습 함수 정의 --> 평균 제곱 오차가 최소화 되는 지점을 경사하강법으로 구함"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# 손실 함수 정의\n",
    "loss = tf.reduce_mean(tf.square(y - y_data))\n",
    "# 경사하강법으로 손실 함수를 최소화 (0.0005는 학습 비율)\n",
    "optimizer = tf.train.GradientDescentOptimizer(0.0005)\n",
    "# 학습 오퍼레이션 정의\n",
    "train = optimizer.minimize(loss)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- 학습 세션 시작\n",
    "Coefficients: [ 0.00364445  0.00183603]\n",
    "Intercept: 0.827988572515"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 [[ 0.011511   0.0354317]] [ 0.00093987]\n",
      "10000 [[ 0.00280043  0.01287403]] [ 0.39743593]\n",
      "20000 [[ 0.00320522  0.00758022]] [ 0.60392851]\n",
      "30000 [[ 0.00341587  0.00482532]] [ 0.7113871]\n",
      "40000 [[ 0.00352549  0.00339164]] [ 0.76730978]\n",
      "50000 [[ 0.00358254  0.00264559]] [ 0.79641074]\n",
      "60000 [[ 0.00361222  0.00225743]] [ 0.81155115]\n",
      "70000 [[ 0.00362768  0.00205532]] [ 0.819435]\n",
      "80000 [[ 0.00363569  0.00195047]] [ 0.82352459]\n",
      "90000 [[ 0.00363992  0.00189511]] [ 0.82568407]\n",
      "100000 [[ 0.00364203  0.00186773]] [ 0.82675219]\n",
      "110000 [[ 0.00364319  0.00185245]] [ 0.82734823]\n",
      "120000 [[ 0.00364356  0.0018477 ]] [ 0.82753348]\n",
      "130000 [[ 0.00364356  0.0018477 ]] [ 0.82753348]\n",
      "140000 [[ 0.00364356  0.0018477 ]] [ 0.82753348]\n",
      "150000 [[ 0.00364356  0.0018477 ]] [ 0.82753348]\n",
      "160000 [[ 0.00364356  0.0018477 ]] [ 0.82753348]\n",
      "170000 [[ 0.00364356  0.0018477 ]] [ 0.82753348]\n",
      "180000 [[ 0.00364356  0.0018477 ]] [ 0.82753348]\n",
      "190000 [[ 0.00364356  0.0018477 ]] [ 0.82753348]\n",
      "200000 [[ 0.00364356  0.0018477 ]] [ 0.82753348]\n"
     ]
    }
   ],
   "source": [
    "# 모든 변수를 초기화.\n",
    "init = tf.initialize_all_variables()\n",
    "\n",
    "# 세션 시작\n",
    "sess = tf.Session()\n",
    "sess.run(init)\n",
    "\n",
    "# 200000번 학습.\n",
    "for step in xrange(0, 200001):\n",
    "    sess.run(train)\n",
    "    if step % 10000 == 0:\n",
    "        print step, sess.run(W), sess.run(b)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 5. Refererence"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- http://scikit-learn.org/stable/modules/linear_model.html\n",
    "- http://radimrehurek.com/data_science_python/\n",
    "- http://nbviewer.ipython.org/github/justmarkham/DAT4/blob/master/notebooks/08_linear_regression.ipynb\n",
    "- https://gist.github.com/haje01/202ac276bace4b25dd3f"
   ]
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "kernelspec": {
   "display_name": "Python [conda env:tensorflow]",
   "language": "python",
   "name": "conda-env-tensorflow-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}