[1]時間序列導論 (1).ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "import statsmodels.api as sm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "index=pd.date_range(start='2018-01-01',end='2020-12-31')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DatetimeIndex(['2018-01-01', '2018-01-02', '2018-01-03', '2018-01-04',\n",
       "               '2018-01-05', '2018-01-06', '2018-01-07', '2018-01-08',\n",
       "               '2018-01-09', '2018-01-10',\n",
       "               ...\n",
       "               '2020-12-22', '2020-12-23', '2020-12-24', '2020-12-25',\n",
       "               '2020-12-26', '2020-12-27', '2020-12-28', '2020-12-29',\n",
       "               '2020-12-30', '2020-12-31'],\n",
       "              dtype='datetime64[ns]', length=1096, freq='D')"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "index"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "import ffn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([-0.63830662,  0.15596197,  2.56654982, ...,  0.07819394,\n",
       "       -0.60527383, -0.73749521])"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.random.randn(1096)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# construct a white noise\n",
    "# white noise\n",
    "# mean=0, variance=sigma^2\n",
    "mu = 0\n",
    "sigma = 1\n",
    "# set random seed\n",
    "#np.random.seed(1234)\n",
    "wn = mu + sigma*np.random.randn(1096)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "z=pd.DataFrame(wn,index=pd.date_range(start='2018-01-01',end='2020-12-31',name='date'),columns=['wn'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# construct a white noise\n",
    "# white noise\n",
    "# mean=0, variance=sigma^2\n",
    "mu = 0\n",
    "sigma = 1\n",
    "# set random seed\n",
    "#np.random.seed(1234)\n",
    "wn = mu + sigma*np.random.randn(100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAakAAAEYCAYAAADmugmLAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOy9eZQkV30m+t1YcqmqrKru6u7qvVu9aAckkEASGEoGDPaY4ZixMd49x2Pw83LGx/Z78+xnz3s+HvCMx8/Ls8dnwAO2x8bYGDAzDEICgUq71NqRWq3e9+rat9xju++PGzciMvLGklmZVVmd9ztHR93ZuURGRtzf/X6/7/f9CKUUEhISEhISvQhlow9AQkJCQkIiCjJISUhISEj0LGSQkpCQkJDoWcggJSEhISHRs5BBSkJCQkKiZyGDlISEhIREz0IGKYm+ASGEEkKORPzbTxBCvrnexxQFQshxQsjERh+HhMRGQwYpiU0JQshvEkIeCD12OuKxjyW9H6X085TS7wu8LjKgpTw+Sgh5lRCiBB77D4SQv07zekrpbZTSyXY/X0LieoEMUhKbFY8BeCchRAUAQshOADqAt4YeO+I+dyOwG0BigJSQkIiGDFISmxXPgQWlO9y/vxvAIwBOhh47SymdCrzufS67WiKE/BdCCAEAQsjPEkKecP/Mg9orhJASIeRH3cd/kBDyMiFkmRDyFCHkzQnH+AcAfpcQoon+kRDyL9203jIhZJIQckvg3y4QQt7n/vnthJDnCSGrhJAZQsgfBZ53j3ssy4SQV2SKUOJ6gwxSEpsSlFIDwLNggQju/x8H8ETosTCL+kEAdwN4C4CPAviA4L35699CKR2ilP4jIeStAD4H4BMAxgB8GsD/JIRkYw7zKwBWAfxs+B8IITcC+AKAXwWwHcADAL5GCMkI3udPAfwppXQYwGEAX3TfYw+ArwP4DwC2AvgNAF8mhGyPOSYJiU0FGaQkNjMehR+QvgcsSD0eeuzR0Gv+I6V0mVJ6CYx53YF0+HkAn6aUPksptSmlfwOgDuCemNdQAL8D4N8LgtmPAvg6pfRblFITwB8CyAO4T/A+JoAjhJBtlNISpfQZ9/GfBPAApfQBSqlDKf0WgOcB/EDK7yQh0fOQQUpiM+MxAO8ihGwBsJ1SehrAUwDucx+7Hc1Majrw5wqAoZSfdQDAr7tptWVCyDKAfWB1p0hQSh8AcAnAx0P/tBvAxcDzHACXAewRvM3PAbgRwBuEkOcIIT8YOKYfCR3TuwDsSvmdJCR6HsJcuYTEJsHTAEbAAsCTAEApXSWETLmPTVFKz3fosy4D+CSl9JNtvPa3AfwDgL8PPDYF4E38L25tbB+Aq+EXu8H3x1yl4EcAfIkQMuYe099SSn++jWOSkNgUkExKYtOCUloFS2/9Gliaj+MJ97G1qPpmABwK/P0vAfwCIeQdhGGQEPIvCCGFFMc5CeBVAD8TePiLAP4FIeS9hBAdwK+DpQ+fCr+eEPKThJDtLttadh+2AfwdgA8RQj5ACFEJITlCyAQhZG8b31dCoichg5TEZsejAHaABSaOx93H1hKk/h8Af+Om0T5KKX0erC715wCWAJyBQBARg98GEzcAACilJ8FqSn8GYB7AhwB8yBWEhPFBAMcJISUwEcXHKKU1SullAB8G8FsA5sCY1f8OeV9LXEcgcuihhISEhESvQu64JCQkJCR6FjJISUhISEj0LGSQkpCQkJDoWcggJSEhISHRs9hUfVLbtm2jBw8ebOu15XIZg4ODnT2gTYB+/N79+J2B/vze/fidgda/9wsvvDBPKd2UdlmbKkgdPHgQzz//fFuvnZycxMTERGcPaBOgH793P35noD+/dz9+Z6D1700IuZj8rN6ETPdJSEhISPQsZJCSkJCQkOhZyCAlISEhIdGzkEFKQkJCQqJnIYOUhISEhETPQgYpCQkJCYmehQxSEhISEhI9CxmkJCSuY/zdMxdxcrq40YchIdE2ZJCSkLiO8btfO46/efrCRh+GhETbkEFKQuI6BaUUpk1xYb680YciIdE2ZJCSkLhOYdpsoOnFhcoGH4mERPuQQUpC4jqF5TgAgKmVKmqmvcFHIyHRHmSQkpC4TmFajElRClxalGxKYnNCBikJiesUpsukAOC8rEtJbFLIIJWApbKBj376aVxdrm70oUhItATT9oOUFE9IbFbIIJWAN6aLOHZ+ESemVjf6UCQkWoLlCicA4IIUT0hsUsgglYBizQTQuCuVkNgMMCSTkrgOIINUAkp1C0DjDS8hsRnAmdRARsWFBRmkJDYnZJBKQLHGgpQZSJ1ISGwGcPZ/ZMcQrq3UUDWkDF1i80EGqQTIdJ/EZkUwSAHAxUXJpiQ2HzYsSBFC9hFCHiGEnCCEHCeE/NuNOpY4+ExKBimJzQXO/o/uKAAALsxL8YTE5sNGMikLwK9TSm8BcA+AXyKE3LqBxyNEkdekLBmkJADLdnD/H07iqy9d3ehDSYTlbqyOukxK1qUkNiM2LEhRSq9RSl90/1wEcALAno06nihwJmU5siYlAVxbqeH8fBknpnu/JYGLfbYMZjA2mJEKP4lNCW2jDwAACCEHAdwJ4FnBv30cwMcBYHx8HJOTk219RqlUauu1F6/WAACnzpzFJL3c1mdvJNr93psZ3fzOry8w8cGpc5cwOTnTlc9oF+Hv/dIs22C9+vKL2KJbeOnMVUxOLm7Q0XUH/Xh9A/31vTc8SBFChgB8GcCvUkqbtqeU0s8A+AwA3HXXXXRiYqKtz5mcnEQ7r/2zE08B80vYu+8AJiZuauuzNxLtfu/NjG5+5+ljl4DnXsXw2A5MTNzZlc9oF+HvXX31GvDii3jH2+/Gq/XzeOLM3HV3LfTj9Q301/feUHUfIUQHC1Cfp5R+ZSOPJQpc3WdICboEgMtLTHzA08C9DNNNUeuqgoNjA5hZraNi9P5xS0gEsZHqPgLgswBOUEr/aKOOIwklqe6TCODSIvNwLG2GIOWKfXSV4OC2QQBytpTE5sNGMql3AvgpAN9LCHnZ/e8HNvB4hJASdIkg+MgLrvrsZfB5Urqq4AY3SEnxhMRmw4bVpCilTwAgG/X5aeA4FCVDBikJH5d5kHLTwL0MnqLWVIIDYwMAgPNShi6xySAdJ2JQMixQtxRlWLIm1e8o1S0slg3vz70O3ieVURUUcjq2DmZweVGOnJHYXJBBKgbBuoNkUhKcRd2wbRClmgVKe3vjwq9ZTWW3+Uhe3xQMUEIiCBmkYhBUcFmODFL9Dl6PunXXMCyHomb29jXBbZF0lWXVh7IaypuAAUpIBCGDVAyCu06Z7pPgTOrW3cMAgGK9t1kJZ1K6wm7zwayKcl06oUtsLsggFQPOpAiR6T4JxqSGcxr2bskD6H0ZumVTqAqBovhMajPU0iQkgpBBKgZcZjya12WQksClxQr2bR3AUJaJYnu9ode0HWiKL6AdlEFKYhNCBqkY8HTf1sGMDFISuLRYwf5AkOrWgn9lqYI/efjUmoUZpk2RUf1bXNakJDYjZJCKAd8pbx3MSFukPofjUFxZqmL/1gEUcjqA7jGpbx6fwZ88fHrNcnHTdqCpPpOS6T6JzQgZpGJQrJlQFYKRvO5ZzEj0J2aLdRiWg31bB1DI8XRfd4QTVZOJG66trC1IWY4DPcCkBrMa6pYjswISmwoySMWgWLMwlNWQ0RR5Y/c5uPx8PdJ9NTdITa/W1vQ+hkWbghQAmfKT2FSQQSoGpZqFQk6Dripy6GGfoyFIuUyqW+q+msek1hakGJPy032FLgdXCYluQAapGKy6TEpXFTk+vs9xabEChQC7R/PQVQU5XemaySxP902vMUixmpSISfVPr1TFsPDE6fmNPgyJNUAGqRgUayaGczp0lch0Xw9grljHSnVjGmivLFawaySPjMZumaGs3jXhRNVg19paa1KmHU73qQD6i0l99aUp/ORnn/U8F3sNNdPGyeniRh9GT0MGqRgUA+m+XgtSvXY864FP/O3z+OTXX9+Qz2Y9Unnv78O57inlah1kUnpI3Qf0V5BaLNcB9G4d7ssvXsGH/uyJnj2+XoAMUjEo1YNBqndqUi9eWsJt//dDmFruL0fruVIdc8X6hnw275HiGMpp66DuW2NNKsSkeC2tnxZEnpKtW72Z4lwoGTBsB2U5MTkSMkjFoFgzUcjprCbVQ8zlzEwJhuXg7Fxpow9lXVE3nQ0xda2ZNmaL9cYgldW6JpyoGmxBnSvV18SYjbDjRKb/mBRPyfaqGTDfkNR79Ph6ATJIRYBSyiToOQ0ZtybVK6MZlqssvz67ujGsYqNQM23UNmBHvODWM7YXst5jhS6m+/jCRSkwswYZumU7Xg0N8NN9/cSk+EaiV5kU35DwFK9EM2SQikDNdGA51Ev3UQrYPSJDX66wNNPsBqW+Ngp1y/Fu6vVExV3UuToO6K5wombayOtM5LCWupRp0ybvPqD3jXE7CZ6S7VWmwoNTXaqHIyGDVAT4xV3I6dDd3Wiv9EotV3mQWlvNYjOBUoq65WzIzVwSBKlCF2tSNdPGwW2DANZWl2LCCf8Wz2gKMpqCUh/VP7x0X68yKVMyqSTIIBUBXnAddpkUgJ6pS630IZPiwWkjmBTvK+I1HcBP93UjBVw1bRxyg9TamFRjkAL6z2SWbzB6lUnx61kyqWjIIBUBvgNjzbwsZdIr/n28JjXXRzUpvshsxI7YZ1Kq99hQVoNDgUoXgmbVsLG9kMVgRl0Tk7Ic2mAwC/Tf4MOiV5PqjXs3DMmkkiGDVAQa0n3ubrRXZOi8JjVX6qMgZW3czcyZx1CwJpXrnlKuZjrI6Sp2juQwvdp+m4FpNTOpwYzW83OwOolV9z7u1SBQ84JUbwbRXoCW/JT+BL+RC4F0X6800HrCiTUakG4m8Ju4ZjKVJSEk4RWdA+9haaxJ+eM6xoc791mW7cCwHeR1FbtG8murSTm0oZkXYNdzv6T7KKV+uq/DTOrLL1zB5Kk5TK9UMbNax4+/Yz9+4T2HW34fT4LeozWzXoBkUhEoNQQpdqP3TE2qaoIQoGzYfbPgBG/i9U7dlARMqpDtzriOmvvd8hkFO0dyuLbc2ZrUYFbrm8bRsmGDlww7zaR+/xsn8PjpOSiEoFgz8eSZ9vwBfQl6b6wtvQgZpCLA0wSFrO5NN+0FJmXaDkp1CwfcxtJ+EU8Eb+L1Fk+U6xZUhSCrNbs3dDrdxxdTxqRymC3WYLV53Vk2haY0B6l+aeYNSu07ubGhlGK1auHH3r4f//iJe/GWfaNte0ry61oyqWjIIBUBTzgRTPdZG1+T4jfD0fECgOsj5ec4FL/5le/i+NRK5HOCN/F6iyfKdRuDGbUhxVjo0rgOHoBzbrrPoe3XHg3bga41pvuGMt1zyug1BFlu2iBQM2187DNP47WrcdciS8kOuynf4ZyO1TaDVFXWpBIhg1QEijULgxkVqkK8PinT2fgLidejju4YAnB9MKn5ch1fOHYZj7wxG/mc4E283jd0uW411KMAP/XXaRGCx6QyjEkB7fdKWbYDPcSkhvqoJrUa+G3SXjMzqzU8c24RL1xcin5fNyAN59k1MJLXGz6rFUjHiWTIIBWBUt30iuO60jsS9BVXfn4jZ1LXQZDifV9xC35wJ7zu6T6jOUgVsq5wosMLfjWQ7tvpBql2eqVsh8KhiKhJ2XB6pDG9mwimNdMyKX7+41KivBTgMam8hpWq2XLPHKU0IJzY+LWlVyGDVAS4bx8An0n1gASdM6mD2wahq2TDXME7CZ7CXI0RITQwqXVO95XqdjOTyqUXThw7v4hvHp9O9VmN6b72mRSvn4b7pIbcXq9KH+zcg79NWibFz3/chmml6jb65/10n+3QlnvmgoFJMqlo9F2QWqmYeOlSNJXn4LOkAPSUBJ0HqS0DOrYPZa8LayT+nVar6ZhUbQOEE0OBRl4AUBWCgYyaqr7znx96Ax//2xfwl4+dS3wu31nndBUjeR05XcF0G8MP+bWaETApoD9MZnmgyetqaqbCz3/c+fGZlJ/uCz6eFsGMgBRORGNDgxQh5HOEkFlCyGvr9Zl/8/QF/Ohnnkk0i+VjOgD0lASd+/aN5jPYPpy7LpjUco8zqXLdarBE4hgKKeWi0j0XFirIago++cAJ/OFDJ2PTQkF1HyEEu0bymGqDSVku629mUt2ppfUi+AZiWyGDekqmUkuT7vNqUnrD/1tV+FUDx9Srtk29gI1mUn8N4IPr+YHLFROG5STKcINMqpck6CsVA4QwddmOQva6GNfhp/tSMql1vqFLdauhR4qDmcz6x/zRTz+N//iNNxqeUzEszBXr+MWJI/jY3fvw54+cwf/7zVORn8W/Wz7DmNvO4VxbNSl+rYq8+4B+YVKsn3DrQMbrP0tC1XDc18YxKe7rqTf8Py4TIPwsc+MUq5sJGxqkKKWPAVhcz8+smuxCSpKMFuuWR+d7Kt1XNTGS16EohAWp6yDdt1JhYpBizG+y0X1S4ZoUAAzldE84Ua5beP7iEo6dX2h4zqXFCgDg0PZB/P5H3oR337gdX335auRnBYUTALBrpM0g5WYKwo4TfZXuq1sYymjI6WpqJuULJ6KvRb52FELpvpaZlCGZVBr0vC0SIeTjAD4OAOPj45icnGzrfUqlEiYnJ3H+Mrvhv/PE0zgwrEY+f6Vcx9LsNUxOLmKhyi6gV4+fwJaVM219fqdw6kINWTiYnJxEZcHAUsXEw995pGFuUBD8e/cyjp9hbHB+tRx5rCfPGN6fv3v8BMaK0b9Dp79zsWpiYWYKk5ONrgJWpYqpIjA5OYnTS8zd4OS1ZTzyyCNeT9ULMywYzF84gUeXToGW6yhX7Mjje/UCW+heOPY0BnUCY8XA9IqJ7zzyCJQEK6jg954ps2v2zKmTmCz7tbALK2xhfOaFl2Fc6fnbPxFxv/XpC3XoxEaluIKySVNdE9+9xM7/9Pxy5POPnzKgK8AzTz4OAJitsHN97KXvQp/VUx/76SU/SF2bm2/pmt0M93Wn0PNXKaX0MwA+AwB33XUXnZiYaOt9JicnMTExgc9feh64NoMbb7sD9x4eEz7XtB0YD34Dtx49hImJo4ytPPptHD56IybecaDdr9IRfO7cMezUTExMvBNT+Uv45zOv4pa33oM9o3nh8/n3Xk+cmS3hk19/HX/xE2/z0lZx+Ofpl4BLU6jZSuSxPlN9AzhzFgCw74bDmPieQ5Hv18nvbFgOrAe/gVuO3ICJiaMN//aPV17AmdkSJibegwtPngfwOsom8Oa778PYEJvie/qxc8BLJ/CR978bIwM6vrX0Kk6sTEce3/FHzgBvnMT77n83spqKy9kL+F/njuP2t92LHcO52GMNfu8zs0Xg8cfwpttvw8RbdnvPuTBfBp6exMGjN2PirXvbPS09g7jf+guXn8c2q4xdY4O4tFjBxMS7E9/vzOPngNdPgGTyke/70OKrGJmb8f59qWwAj30Luw8ewcQ7b0h97OrpOeDZY8hoCvJDI5iYuC/1azfivt4obHRNat3BC6NxRfqguSwAryGyJ/qkKgZG3fTCDnecea+JJx46Po1HTs7hylIl1fO5us+wnUgpbt2yPVui9ewpqQjMZTmCwonXpla9x8/Nl70/X1wsYySvY2SAi3CU2FaGmmmDEL8OykfWt+o6YbjuKJk+TveV6hYKOR05XYWRuiaVLEFfrZleIy/grxPtpvu2DOhS3ReDvgtSvJchriblm8u6C0sv9UlVTYy6C96OYbaA9Zo10uvugp02mCwHfouozUPNdFDI6VBI93pK5op1fPS/Pt1QAxKZy3IM5XyLoeNTq96gwnNzJe85FxcqODA24P1dU0hsbbNq2J6yD/CL8q2q8SzXHSXs3ce/R6kPZkoVa0zwktWU1NdM2poU/10AQFMVDGW1toUTo/mMtEWKwUZL0L8A4GkANxFCrhBCfq7bn1lJuVMC/Bu6pyToFTPApFj6p9dcJ7gHX9rd4WrVhOrW1KJu9LplI6cryOlq14QTr02t4NiFRbxyZdl7zJvKK1T36SgZFmqmjdMzRXzfbTuR0RScnfOZ1KXFCvZv9YOUrimePFyEqml7ogn+GUCy0CcMT92nNd7iOV2BQvqDSXGFblZXWu6TqplOpLHvas3yZOccwzmt5T4pHjhHB3TZzBuDjVb3/RildBelVKeU7qWUfrbbn1l10zdp0n3D4XTfBgcp26FYrZkYGcgAALYNZUBIbwWpUt3ChQWW5kurWFquGJ67QtTvUjcdZDUFeV3tmlyXB4Klsi/SEE3l5ShkNVAKvHhxCZZD8ea9Izg4NuAxKct2cHWp2sCkdIXAsJ3IXqmqaSMXCFI8rSTaVMVZG3HWr4cENYSQpv6uzYTZ1RquLqdrbmZBSkdOU1MHgeDzoiYYF6umtzZwDOf1ttN9owO6tEWKQd+l+/hOKY6al+uNdQhFIYlpmvVAsWaCUnhMSlMVjA1mMNdDMvQ3rvm1mXqK8+U4FCtV02MbUQyXMSnVZVLd+R34IrNU8Rcb0VReDm6N9Mx51kVx++4RHN4+hHMuk5parsFyKA5sHfRew9sZoprJ66bTIDbxenBCwXtquYqb//2D+G6A9QURxaT4d9msQepXvvASfvUfXkr1XNaQ3yKTCrD0YkTKj9WkQkwq37oTetXdxG0ZyEgmFYO+C1JeTSrO3cDyrWk4kgre6wEuMOA1KQDYXsj1VEPv68EglYJJlQwLDgX2bWFBKupGr7lMKqsrXWNS3Oh2qeIzKR6kBgSOE7xg/sy5BRRyGvZtzePQ9kFcXKzAsBxcWGDBan+wJqXG1zfD6T4eCMObqgsLZRiWgzOzJYjgOU4IWhMGs5vTCX2hVMdzFxaxEGC6UTAsB3XLQSGrIaupsByaai5XsDYkCuR8llSwJgW054TON8wjecmk4tB3QaqaQjjBL9Sc7p8eXSWpFULdgmeJFAhSrKG3d4LU8auBIJUimPDAsG8rk9BHpvtcJpVvoTGzVazEpPuETMp97OVLy7ht9zAIITi8fQi2Q3FpsYKLbiNvQ7rPrW9GjX3hwgn/+QoGMmqTke2K53coPl9GhOMEsHkHHz5ycg4OTdfMzb9fIad593GaQBB0gRD5MnqzpPKhdF8bM6Vqpl9nNSynL5zp20FfBSnDcmC5F0KccKJmNjOpjKZseLpv2d3hj+Qz3mPbC9mekqC/fm0VN7gqtzSLwrIXpOLTfZxJ5XS1YSHpJJarzUyKM29hTcplOYbt4LbdIwCAQ9vZnK+zcyVcWigjoykYL/j9Tf4AzYggZdrI6o235XBObwreSVZSnEmJglQhtzmD1LdPzABIJ/rgQX0opyOrsd8ubZDirQ6ic+T59uXC6T6t9XSfuyHh64xkU2L0VZAKLm7xZqZukNL8hUlT4lVZ64GVCCY1X6r3xC7MtB2cnCnizn2jAJCKefLvtGskD00hkTd6kEl1S64rqkmVQvXJIAqBher2PcMAmP0RAJybK+PiAlP2KYGUGzd8tSJ+r1oo3cc+p1nezI81qljve/cJ0n2ZzZfuq1s2Hjs1BwCpNinBXkefSSW/rmba2OY2YguDFHdAzzen+4p1K9G4Ogie2m3l+PoR/RWkjHRBiu9ogjtaXdt44YRXk8o3BinLoVisJOfpu42zcyUYloM797MglYpJuUMcRwd0VnyO6ZNiTErpmgRdlO4r1y1oCvF210EEU4CcSQ3ndGwvZHFuroRLixUcCMjPgWQfyKppN7l0DOf1piK+x6QSg5Q43RelXOtVPHNuEWXDxlv3j8K0aeIGKBikOJNKs7mpGrbXQC1K93mzpMLqPq+fLT2bqpo2chm1pePrR/RVkOLuAcOCnWkQvOs/uDDpqrLhfVI8SI0Eg5RrldML4gnexHvn/i0A0u0Mg99JxBg4OJPKrocEPSScGMxqXnNtEFzUkNMVr5EXAA5tG8QZN0gFRRNAoCYVJZwwmpmU6HpdTmRS0em+oay66dJ93z4xg7yu4n23jgNIrkvxYFHI6gGnkuTrpmoGglQLTIr/vZWG3prRyKSkwk+MPgtS7CLYOZJz5dzRKZecpjYsTBm1B2pSVQOFrOYpxABg3HWdmGnBdaJm2i2lJdLi9alVZDUFN+1ko+3TqPv4IjuS1zGc0yN3orVAn1S3HKP5sSxXTe/8lOo2BiP8B4dcxd/NO4cbfpPDO4Zw/OoqKoaNg2ODDa/hQSNKaVYL9UkBLK0YXZMSny/PcUKQ7htya1KtjjvfKFBK8fDrM3jX0W3Y6vYIlo34YNAonEjPVFi6L9PwHkFE1aTacULn6b5Wamb9iL4KUnynMj6cg0OBcsRurGY6Dco+oDck6CsV0/OA49gzynbqV1I2OFJK8b4/ehR/+XjylNhW8fq1Vdy8swBdVZBR0/WmrFRNT+E0nNcihQB+n5TSPeFExURGVUCpvxhFjekAWP/c9kLWS29yHNo26LHuMJPiNkVRrLwW6pMCWFE+LChZ9ZiU+HzxdFhUus926KZZFE9cK2JqpYb33zLunZukUe38fA3lNJ9JpbhuqoaNwYwWOXXZmyXVpO5zWwVaTPflM5JJJaGvghS/sMfdFFl0T07zblZXe6AmFfDt49hRyEJXCa4upQtSc8U6rixVcXkxnflrWlBKcXxqFbe6tZmspqQSTixXDG8XGiXjpZQ2MKlu3MyG5aBq2l5Q4TW+shEdpADgS79wL379+25qeOywq/AD0FSTymiucEKw4bFsJm9uTvex8xJkPkk1KStinhQQ9O/rfMqvG/PNHj4xA0KA+2/e4U1IrrTApLIp1XOUUtQsBzldjWx4jlb3tW5fVTX8BvU0x9ev6NMgxVJkkUV690INQlfTLbrdxHLFwGhAfg6w3fzu0Xxqx/EzrmVPp9nI1EoNK1UTt+5mKjfW5Z+iT6pqet+pEOF/xllHljtOmHbHU1V80efpOS73j5rKy3FgbLDp33mQUgiwd4uYSYk2PHx6bJjFF3I6LIc2/GbLCX1SZhyTynTHCf2Zcwt4x6e+jYsL5eQnt4AnzszjzXtGsL2QxUBKJrVaM5HRFGQ11WNSSZsb06awHYp8RsVQTvMGWoreN7w+tJPuq3npPsmk4tBXQYpfBDs9JhXVk2M3qbyhq/gAACAASURBVLnWq0+qYliR3mTL1eZ0HwDsGc2n9jPj5qedviG4aOLWXW6Q0tTUfVL8O7GaVPNvwmsJvE+K0s6b/fpBymVSZfb3St0W9kjFYc+WPDKagl0j7P9B6DGOE1wM0MSkBP59/HijZM98Mq/IcYILPlp1Vk/CyekiKAVmOizimV6p4aArTBlwNwTJwgkLhawvbAGSmQrfBOR0FYUIVw6R2wQQYFKtpvsCTEoGKTH6KkitJd2nKSSyt6WT+Mxj5/DhP39S+G8rAQf0IPaM5lOn+866NjpJO9FW4Y1JdxeTrJa+JuWl+/I6KobdtBmoB2yqWimCtwIvSLnHz2XopZiaVBRUhTlP8J6pIHSvT0rApARN5EDAv889Rsc1GuYMTiQ2MW0HukrEqsQuzZSacjdKnWbpi2UDWwcZ2+ZMKlE44TqgAwhIvOOPi/97XleZK4ewJmU21aMAYDCjQlVIa8IJo7EmJdN9YvRZkGIXnRekYhy3RcKJ9Uj3XVuuYb5Ub9odU0qFNSmApZRmi/WGm7Bm2vjZvzqGcyuNN+ZZnu7rcJDigYQXtjOaAiN1uo8zKfEOv97ApLqTGllx+7W4W8ZSoCYVl+6Lwv/3sTvwex++velxLaZPii/uoj4pwL9ei3ULlPouHaKMgGU7TbOkOLzBhwkLfavg4p1qB9+3Ztoo1S2MhYJUsnDC9Jqts2mZFGeyGSW2JiViUoSQxNaWMJhAS00dRPsVfRWkqgEJOhBjwWMJhBPrlO7jC1F4N1py0zrhmhTA0ksAcC0wrO+N6SImT87h2anG78iZVLs3xFdfuop/K3ChDqbk+P/Tpvt44I2anRRkUvkupUZ854scMqriCyfqltBcNglHxwseKwsirk8quJMPouApx9wxM+6x7nf9DkW7d9OmQtEEwPqkgM4PPuwGk1p0Ge2Y6wLBf4tKAgvkAw8BpJZ4VwPnfyjXrKjk7xvukeKIa0YPIyiS4UG0JpmUEP0VpEwbmkKwxe21iE33aY0LRWadJOh8wQmrl7ym14iaFICGlN/pmSIA4OyKf+GX6xam3EDW7kLy2Ok5PPz6TNPjdctGRlO89FJWS+5nqls2qqbdkO4DmjcP4ZrUWo4/CiuBpuItgzqWyybqlg3Tpt6i3gnEOU5E1qRCwdvzO+TO8ZHpPvHtPZRl7ydKZ60FPEh1MpXMg1Q43VdJ+P3Z6PjGmlTSxqapJiVghKu15llSHMO59DOleEDKZxQ/iEomJURfBamKmwPOuFLmOAue5nTf+kjQ+TFVQrtcz7dPsIvb6zKpoMKPj3C4uOp4x31+nokmBjJq2wvJcsUU7kj5UEKONOo+r5HX3TRE9ZqImVSna1K8/0XHloEMFitG7FTeduE38wqEE3yRFPRJAT6T4ueNy+XFTCo6SHEhSCdrUobleG78nUwlz5fYe/IG26zGJguH748w+MBDgG0wCUlmUjXDv854TSqsIl2tRjOpkRZmSgU3JLImFY++ClJByxnmWhyt7hNJ0NclSLnHFA4i/iyp5nTfzpEcFIIGhd8pl0mZDlNdAX496rbdw22ny5YqhnA2T91yvB0hwBaGJAXeSsjmKarXJMikeGqk0zW1laqJwYwKXVWwZSCD5YrRNPyyE+BqO9G5ERkbA81MigclvybVvDBaNhW6TQC+BL2TfVIzqzXw9byTv43PpFi6jxCCwYyWsialea/JakoiU+F2WzzdZwkanhmTikr3RTejN31WgLV5QVQyKSH6KkhVTNtLF4jGH3BEBan1EE54TCqc7gsYsYahqwp2Duca032zJbzVdUJ46TKb3np2tgSFMBufdhcSHizDN289JNvP6kpium85xA4LUcIJkbqvw/59K1XT2wBsHcxgsWzEzpJqF1ySHsekwsKJrMYcPIphJrU1Ot1n2A4yEUxKUQgGMmpHmdSVwLXXjZoUT/cB7PzENfNSShvSfUC6lgg+8TmfUT35evBarJk2DKt5lhTHSAsj5IO/NQ+isiYlRl8FKSb5ZBfYcF7ckwOwfHF4pg/rk+puTcpxqG/HY4jTfVG7uL1bBryFoly3cGWpivtv2oHhDBvKB7BG3v1bBzCS19tuiOWqt6YgZTWmSNMsCiuhScNRvSYek9IVjwl3ete5UjW8zx8d0LFcMbvKpMQ1KXeRDG2QCCHuLp17C7LfYPdIPlL2HMekgM6PkJ8KsPhO1qTmSwZ0lTTUgQaz8UyqbNhwKEJBKjn9HBZOAI0pUc9cNopJtTD4MFx/zHXJSeV6QH8FKdPymFSUu4HtsDEA4ZQL65Pq7k6n7I5SB5plvPxmGYoo2u7Z4jf08rTe0fECbhhR8fLlJfb4bBlHdgwhn1HhtNEQazvUWxDDN1TdshvSfWkWBc6keLpvKKOBkBh1n6Z2TzhRNTHi7pC3DmawVDE8x4FOCidSSdD15s8rBBbAlSp3PVAiZc9xNSmABanlSroFNQ14kBrJ6x1dbBfLdWwdzDT0e+X1eCbFBSFcIALwIJC+mVeUEuXnuRAlnHDHwKf5/uHfmqUjJZMSoa+CVCVYk4rY9QRTS0Fwg9luOkcH89nheT9cLjwgWMAAJp64tlKFaTs4NcOD1BAOjyo4O1fGUtnA+fkyDm8f8s5Bqym/larp1R1ETCrIPjMpvPt8MQhL5SgKwVC2Oa8vYlLdaOYd8ZhUBg5lTgcA2pKgR4Gn4ESN4V6dItN8Ww4HJNGr7rESQiJTTKZDG5zZw7jr4BY8cnK2Jff8OEytVLFtKIMtA3rH0328HsUxmI0X/nhjOlpkUjXDT8GJXDmixnRwtOI6ERbJ5Lo4gmazo6+CFO/wBqKLnHzxC6v7eC2hmym/YNAMS2zLdQuDGbVhymsQe0bz3sJ6eraIjKrgwNYBHBph3/frr16DYTssSGXaYyPBOUtNTCqs7kvRJ7VSMUBI42IiqhXWzSCT6p5wggeprYPs/1wt2cmaFE/BicbH10wbCoGwlhTswQm7dAgl6JaDTEy675fvPwrbofiLR860/B1mizU8c26h4bErS1XsHs0jp7evHBVhvmR4yj6OfEaLnGAA+Ju9YNahJSalKShwmX4Dk0pK97kqzBQNvbVwuk+T6b4o9FeQCgsnqs0zpaKsafwmzO5R8oYgVW9O98XVRnhD79XlKk7PlHBo+yA0VcENI+wn/sqLVwAAh3cMts2klgNBKhyAwg3QaWpSy273fjDwDuf1pps8OCl5PYQT/P+XF1kKqys1KQGT4upTkZURGwjp90nxIBXFpCwn2nECYPL1H7lrL75w7HJDPSkJD752DR/448fwsc88g2sr/uumlqvYPZLHQKazi23QEoljMKPGulrwwDLcRk0qoyrQVMULcKXARGQe/EYihBOcSaURTzSl+/R0ze/9iL4KUhUjEKTyzFk6vLvyg1SUMWj3LqTgxR3eKSa5cXO37StLVZyeLeLoOBs8OKgTHN4+iBdd8cShbUNt13WWyv7xpWFStkCqHsSKwOZpWFAr9CXoQcfozv0ONdNGzXR8JuUGKc6kWjWYjQMhJLLnripQlXIEzXcbraTEaWvDptAFI++D+KX7j4CC4s9TsKly3cL/8aVX8At/96LXf/T8BVbrpJRiarmGPVvyrvKuc0FqoVTHWCjdl8+oTenwIPx0n39tpVGbVg3bS1kPClw5kpjUSBvpPp7VkEwqGn0VpPj8FiDQexKxIIaFEzxIdXOEfDD9KBJOxO3od7lWT2dmS7i8WMXRHf5Mozv2sXHuY4MZbBnMeIG6VSa1FMOkwsIJnh6N2x0uCwxzC4JFt27Z0FUCVWGGqTld6egN7S0+7rFwR5LLS1XoKmn4Xp2ArirC4B0bpCLTfZpw8KFlO9AjUsMce7cM4GN378cXn7ucOF/sDx58A1964Qp++f4jeOhX3428ruKFiyxILVdMVE0bu0fzyOtqx1KxNdNG2bAxNhRmUlrsBssXTgTSfVpyzadu+TXrgsCVI7EmFWHrJQI/RznJpBLRN0GKUoqK0ajuAwSNoxHCiUzMiIUo1EwbP/25Y/iTh0+lej4/loyqNDGpcsLIiJyuYnshi0dPzQEAbhwPBCm3X4rPOWq3JhVUgzWr+5qZFIBY8cRy1Wy64UVTaGtmo9oybvAhpa2LW4Ij7AFgi1uTmivWO5rq49AUEundF+6R4ihkNdRMB4blYCVw3iJrUgnqPo5fuv8IFIXgLybPRj6nZtr455eu4kNv2Y3f+MBNyGdU3Ll/FM9dWATgN5HvGc0hnxBAWoGoRwpAYo8Xv34ahBMpmZTHbHQFqkIa031VCxlVaRrjw+E5g6QIUmGfxqyWXDPrV/RNkLIo4FBfqRXdk8MunnCfVFzBWwRKKX7rn1/FY6fm8PfPXoKTYswHP5bthWxzTcqwPFlsFPZuyePENTbX6ciOgvf4nfvcILWDGZ62W5OKZ1KNgyLTTENdDdSBOITCCctu+D1yMbv1P/rWKfz+sdYUa+EgNZTVvBpk0jlvB1GzyYKOKGHw63W5aqBYsxqmGRsC2XNSnxTHzpEc7j00hteurkQ+56Hj01itWfjoXfu8x+46sAUnrq2iVPfnnzEmpXSMSS2UXHPZpiCloW45wjlaAHBuvoRCTmv47XJpmnlN//wTwpSmYSY1nNeENUMgmJ1JFk5UTRuqQrzrLKcnO2L0KxKDFCFknBDyWULIN9y/30oI+bnuH1pnwdd8X4IuVuLUPXWfON2Xtlfqs0+cx1devIo79o1itljHa1PRiwDHSpXNCCrkmpsVk9J9gG80q6vEG94HADftLOAte0fwnhu3A0D7NakYJhUeFJn10n3Rn8FGxzd+p+G8jlLdagjqbHS8/3swuW7z70ApxT+/dBWXi63tSMO+iIQQL3h2sh7FoSmK0HGiZjaPjufgu3TesB0UTgDNu/c4x4kwxty+sCh86YUr2DOax72HxrzH7jq4FQ5ljeJTgSA1kNGEPUzMNqk1hrtQZr594XSfP65DHAyeODOPew6NNQhysilSxFWzcaPFGp4ba1JR9SgA7tgNJWW6z2kQyaQdEtqPSHMV/zWAhwDsdv9+CsCvduuAuoW6uyjkA8IJIJpJRdakrOQb7dFTc/jUAyfw/bfvxGd/5i4oBPj2idnE161W2Q5Z1FFfqtvJQcpV+B3aNtTQI6OrCv7HL78LH7x9FwD/Jm+1rrNUNrDFFTqk6ZMSPY/DcRuDw6NHhnMaKEXD6G4RkxId+/n5Mq4sVVGz0NKCuBzyEAR88URX0n1xwonIdB+XxbOAEHbpCCvK0jIpANgymPGGPIZxdbmKJ87M44fftrdh0b9z/ygUAjx3YRFTy1VkNQVjgxmh1Ht6pYZ7f//b+KcXrqQ6Hg7OpMJ9UgPZ6EzApYUKLi9W8a4j2xoeTxMEaiEmy4JUo7qvEFGP4hhOaY0Urj92us56PSFNkNpGKf0iAAcAKKUWgE13Nvn1HJSgA83U3K9Jhfuk0kvQf+srr+LojgL+8EfegrGhLN66fwu+/UbzeIswVl1TzAGBN1m5biU6H3CF39FAPUoEfiO2qsJaqhjewMhgasKyWeql0XGC/TmqJlVy3TWa1X18XEeQtTXWpKJuaF6Po2hWR8YhnO4D/OPqZI8UR0ZVhBL0mmkjr0fVO9jxcIFDE5MKbbbS1qQAVvMpG7bwnH75hSugFPjht+1teLyQ03HTzmG8cHGJKftG8yCEeQIattMgDJlZrcGhwF8+dq6lzYM/S0rMpES/8ZNn5wEA7zwy1vB4OibVWBMcymlNfVJRYzo4RlLOlGL1x8aNl2RSYqS5isuEkDGwex+EkHsAJOeuUoAQ8kFCyElCyBlCyP/ZifeMgsekuHonSjiRkO5LClKrNRNXl6v4obfu8Xbh33vLDrx2ddVzMACAp87ON81lWnUL4uFRGrZDUTWTmdReN91343gh9nlrEU5wFWHwhvL6mFpI9/Gd+4hAOAE0pmHDTCpKOMGDFNDarKSVarNqixfru1GT0lQirG3Gq/t4uq8xSPFFM7x7byVI8YActklyHIp/euEy7js85jmuB3H3wS146dISLi1WsNu99vKCVDJf6E/PlvD46flUxwQAC2Xm21cIXffe4ENBuu/JM/PYUch6IiEOzqTigmSwJgUgoiaVwKRyWjomFWJtWU0yqSikuYp/DcD/BHCYEPIkgP8O4FfW+sGEEBXAfwHw/QBuBfBjhJBb1/q+UeCp5XzAhiSjKdHpvogglSRBvzjPFpFgTeh9t4wDgMemZldr+MTfvoBPPXCi4bUrbs57INM4cI3/OWlXf+POAnSV4G0HtsQ+z+s1aoNJ7SjkmsYKiM6ZF6QiFEtTyyxg7xrJNzwuag2oNzEptSnA1kwbz5xbwI4CSw0F0zRJWKmaKGQ1qIF01mgX0326qghrm3HCCd7zwxuMm2tSjQt23GTeMHhqczGU8nvm/AIuL1YbBBNBvO3AFpQNG69NrWD3KNu8iDZAXG2nKgSfe/J8qmMC/B6psFAhaoS841A8dXYB7zqyrek1nto05v4N9kkBzSa8q1UrtiYFsOslrofL+6xQQMzpqnAEjkSKIEUpfRHAewDcB+ATAG6jlH63A5/9dgBnKKXnKKUGgH8A8OEOvK8QnEkNBOg8a4QU+8RFN/PGpysuLLDBgsHR4Ud3DGHf1jy+49alfvd/vY5izcJ0qJjMVVsDmUb1Wlo37j2jebz4O+/HO0P5+DAIIayfpYWdG6WU9TUN6k1jBURMKqkm5cmWtzQGKdEIeTGTanzfZ88vomY6+IE3sbpblMO9CKtVs2niMbdG6qS5LIemKjAiRnVEN/Oy3/4yZ1JJNSnHifXuC2KLyxqXQ+KJB1+bxkBGxQdu2yl83d0HtwIAKEUzkzKamdQP3bkHkyfnvIGcSRC5TQBBJtV4/b4xXcRi2cB9guvfcyqJkXkH+6SAxiBVNWwsVwzvuohCWgGEqCbFjkEGqTASt4mEkJ8OPfRWQggopf99jZ+9B8DlwN+vAHiH4PM/DuDjADA+Po7Jycm2Pmy1XANAcPyVl1A8zy4OjRo4e+kqJid9H7I3TrMb9ZknH4cS2I1dWGE3xIsvvwI6FX3aHjnLXn/x+AuYfsN//c0FE5OnZvEH//Awvv7dOrblCearNh54eBKDOnveYqmK1YVplAnBatX0vutUiV24F8+ewmTlXEvfu1QqCc+ZRmycuXAZk5PJgg4AqFkUhu1g6dplqHBw9sIlTE4yZjhdZsd37ox/fJdW3fP1yndBppvP15Nn2Hk6/coxXAzs+Gcr7L2ee/lVZObeAAAsLFegmYr3PZYX6lgu2g3f6+9P1KErwHZjmr3/sRewci4dCzp7pQbVpg3vtzDFFv356SlMTqZPUaVBtVSFUUbT71KpW5i7dhWTk3NNr3EoBQFwxa1JvfLc09AV4hnVvnLiFMbGDUxOToJSCtOmmLp8CZOT04nHc9VVQz7x3Mswrvjn7NWzNYxlKZ596vHI127NESzWKIrTFzE5OYVz02xRf+ypZ7GvwBbely6yc3lnfgFfVYBP/dOT+OnbspHvyXH+WhV5rfk8cfXmsRdfwS1DNe/fHzzPPkeZO4XJyUYXjYuX2L898tjjGM2Kg3epamB+ZspbD5bn6lgpW5icnMQrcxYshyJfvBJ7TleXalhadRLXqbmFKgoZ4j3vknuOvv3o4xjOJDPgqPv6ekSau/juwJ9zAN4L4EWwtN9aIPolmraXlNLPAPgMANx11110YmKirQ978vPfAmDg3ffd47GcncefRC6vY2Li7d7znq6eQObCBXzv/fc3vP7EtVXg6cdx8y23YcLdrYvwtdlXsHN4Hh94b+PrtT3z+NZnn8V/e83E0R1D+N8mDuPXvvgKjr75Ltw4XmB1pwcfwK1HbgAAPHjhNN797vdAUQhevrwMPPEk3n7nmzBx83hL33tychKiczb8zHewZftWTEzckep9rixVgIcfwdvedDMenT6FbTt2YGLizQCAN6ZXgccfxx1v8s/N2bkS8NSjOHLTLZi4Y0/T+z0w/wq2F+bwfaHztFQ2gMe+hd0Hj2DinexcaM89gr27RjExcScA4JtLr+Lk6nTD9/q9FyZx75EteO+7bsZ/fv5xHLop/ncK4s9OPIU9wwomJu7xHlsoXME/nHwFt954CBMTR1K9T1p8+tQzsBwHExP3eY9ZtgP7wW/gpiM3YGLiqPB1hUcfwmrNQk5X8P7v9c9b/pEHMbZzL4YGZzExMcHqpg99A0cORb9XEHPFOv6vJx/GroNHMHHvQe/xPz7+JA6OaJiYaNo7enjntZfwtVem8N577sR9R7aBnpwFXn4Ot73lTrx1P0s7v/ad08CJU/jhD74Hr9Rew9deuYY/+tf3CadMB/E7x76DN+/b4v3uHJcWKsCTj+DQ0ZsxVDzjXQd/ff4YDm2v4CMfnGh6r/kXrgCvv4K33X2PsL5GKYXx0AO48dBBTEzcBAB4yTyFhy6y+/A7XzuOvH4F/+bD90eyXYDd/1eq88J7LohPvvgo9uwYwsTE2wAAM89dAk68ire9/R6vlSQOUff19Yg06b5fCfz38wDuBBB/daXDFQDBZPdeAFMdeF8hwjUpgJuZhh23mwceAoF0X0JT7oWFMg6MNd8Eb79hK4ayGgzLwac+8ibvRrnmiimKAcsV3pvD03Feuq+DRfx8i0ag3LdvdCDjpjT819YFKdJsQrpvarnmpYiCKAj618SOE/77Xl6s4OxcGe+5cbuXLmxVOBEWcPjCiW6k+5odJ3j6NKomBfipvfCxhk1mubgnybuPgwsnFsuN98J8sY7tQ/GM555DLOW3373mvVEqgVRcsW4hoynIaip+5r6DqJo2vvFaMsNbLDWP6QD8ezgonDAsB8fOLzZJzzmShDyG7cChaOqTApgS9TtvzOKdR8ZiAxTgNuWmTPc1Cie6M8zzekA7q14FQPL2LBnPAThKCLkBwFUAHwPw4x14XyFEQaqQ0zy1FIdodDwQsEVKuAAvLpQ9oUTD6zUFv3j/YTgOxd0Ht3pS4hk3SPFFmU/NBVyXiUBevJNF/FY91niz55aBjCsB98+D59IhkKDH1aRu3TXc9LimskF+i24jJ3uPcJ+U0lBPe+w0S4+958bt3sKSRgbMIQpSfOHuhnAiIxBOeF5uMUGRFe2rzb1l+cbBhzwAagnefRy6qqCQ0xoaeimlmCvVsb0QH6R+9K59uHXXsNf+IBI1lGqWp9C7eecwFAJcS3Bej/LtA/wG6+BnvHx5GRXDxn2H44NUVE2qJpiKzJ3Qv3t5BVeWqviF9xyOPWYg/YTdWqgnjm/wpDVSM9LUpL4GPw2ngCnxvrjWD6aUWoSQXwZrFFYBfI5Senyt7xsFXqgODg0UCyfsJtEEAOgp+qSKNRPzJQMHxgaF//6LgbTRjmF280+7Q+f80dQa+NrCFy6+Y+xokMq0JpzgC9jWQb2ZScUIJ0R9Uo5DcXW5ivffKk5djg/nMLPqB6layAkgr6uwHerJrJ86s4DdIzkc3j7oTTZOOxqdUtZUHBZOHN4xhLfsG8VbXEupToJJ0CNGxMSwH84ym2T7OTGTyqRkUgBjjkF132rNgmE5iUFKUxXcud9Xk0ZJ0PmCryoEWwczmCtFO1wATH4ONFsiAX6jfdmwAfdUPH12AYSgwRWj4TWeTZf4mg+7kgM+k/raKyzBc//NO2KPGUCTqCgKTRL00PEtlg388bdO4Tc+cFPT791vSLPq/WHgzxaAi5TS1lrHI0ApfQDAA514ryTUbXizYjjY4MPmPqmw2wSQrk/q4gJjRzdsa073hZHVVIwNZrx0X9CJ23EVf1zKyq1ZOmnRk9fVJjVXHHgPzaiASflBSiBBFywKC2UDhuVE5t7Hh3OYKbLzQil1HdYbGx8BtrDrqoJTM0XctmcEhBCoBMip6dN93LRVtPD/j196Z6r3aBW6qsAMMynBIhlG0FQ2iJG87m12AHiWS3HzpMLYMtBojTRXZJuEpCAVhkiCXqo1jpnZNpTFfKne9NogFktic1mATXAe4DOl3FNxcaGM3SP5ps0GR1JLRHi+E+AzqW+8dg03jRdS1Ypy7gbKsqPVlZRSQbqvkUk9fnoOf/vMRewoZPEr7+1E4mrzIk1N6tHAf092KkCtN+o2bWJIInPOmiV2ovb7pKJrUlx+HsWkwmCMgS0uQdcDLrGtmmyh5TWpTroftCpB5wvYaF7EpJpdOuIWhaAhqQg7hrNeGtS0aVOtIBvYrVu2gwsL5YbmzZxGUkvQRW4T3YauNhvMctYcW5PKiWtSYSseryaVsk8KYMEgGKR4ENmWUJMKQyRBL9ZbD1Lznm+f+PMHMmqD48RMsYbx4ehjTRqW6Y/OaOyTAhirnLh5e+zx+p/jBpsYNsXrX/mGdF/j8fF75K+futD3Tb6RQYoQUiSErAr+KxJCVtfzIDsBw/b7KzhE/n010xYyqUwKJnVhngepZCYFsBlQHpMSCCc4kyrXLSgkfgFrFa2m+5YrzLJJU5VmJmU2MylCCDIRI+SvLvHRDtFMarZYh+NQLwAGmRQ/D3XTwaXFCkyb4khgflZeS5/u25ggRZoMZkU7+TC460TYSmokJADyg1R6JjU6oDcMtew0kwqOzdg2lPHePwqLEQ7oHAMZrSEQTq/UPMsuEXhNM4pJiUb0BAPr/Tclp/oA/x6ICyy8/tXQJ6X51zQAXFuugRCWdfhSi56H1xsir2JKaYFSOiz4r0Apba549zjqNm1o5AXEDtK1SHVf8qiOCwsVjA9nm4JhFMZHfCbFa2PDOQ15vdH2pVRnYzqiRgS0AzbuIn2RdqlieMMAw0yK3+Dh85ZVxSO7pyIaeTl2DudgORSLFcOfyitofKyaNs7OsY3B4e0+e81rpMGgNg4bEaQ0AZPyalIx6b5CFJPKaSjWLS9NzIUTrQSprQONNSkvSLXIpPhiWwk184qYVJxFkTdLSiCcAJpnSs2u1uODKoRfcgAAIABJREFUVIKQpyZgsvyYC1kt0cWFI01TrmhD4gVR936ZWq7ipnE2veC/PX4ucixJPyD1VUwI2UEI2c//6+ZBdQN1uznfPyLo1o9S93HLnPiaVDl1qg8Adg3nsFg2UDNtrNZMKITJzMPqpTRjOlqFl9NPiaWK6TmgZyOZVChI6YpQOHF1uYqhrBZp1snTNjOrtVgmVTNt1o8F4ND2RiZVTKnu43W59QxSGVVplqCnYVJRwom8DkoBrgHi12haF3SAuU5UTd9kdq5Uh66Sls+LojRPTg4KJwDGzmqmE2sCPF+uI6MqTb59HAOBTEC5bqFYtzwxkgi+ei69cIKzv++5cVvqgB+sl0bB/yxxnRUAplaYae8n3nMYFxYq+ObxZMn+9Yo086T+JSHkNIDzAB4FcAHAN7p8XB2HYdOmBUAUpMLD+zgIIZHu1Rzn5yu4oYUgNe6atc6u1j1zWUUh3o1S9oJU/FTedsBrUmldqZfKhtd8GbZ+EQknRM/juLJU9VyzRdgx7J8XkeFv0OLm7GwJ2wvZhsU0r5HUwglPRdYiY1gLNIU0ebSlS/dF16QAoGJyJuWq+1phUm5ajdel5orMN09JKWMPIjxTigkn/GPmda75mJQf65HKRF4jAxnNY1Kz7vvsXAOTEp3/4ZyOD9w2jp+652Dk+0Z9TmyQErC2cF/h1HIVu0fz+MBtO3FgbAD/tUUH+esJaa7i3wNwD4BTlNIbwBwnnuzqUXUBLTGpCOmuHuFeDbDd4nypjgMplH0c/KaaXq155rKA37RbDaT7Oj0yIp9R4dBkw1wOlu5jxxeeIipiO/zvwprUcjUy1QfAS9tEMalguu/MXKkh1Qe4QSplum/BLeBH1T66AU3ApKqCOkUYHpMS1KQAoOLK2rlVUktMKmQyO5+iRyoK+UAquW7ZMGynsSblvm+ceGKhbAh7pDiCkwJ4yjw+3ZfApDzhhH/+FYXg0z91F+49LJa1Cz+nhXSfeONlo1y3sFI1sWs0B1Uh+PnvOYRXLi/j2PnF1MdxPSFNkDIppQsAFEKIQil9BEA6L50eghFTk1qpJKf7ANbBH5Xu46KJVpgUH3txbaWK1ZrlFcb5DisonOh0us+7KVLWpZYrZiyTyqhK0647o4lHYk8tV2PlvLwOMpPApKqGjbOzpQbRBMDTfemC1HzJQCGnJToJdBIZlcB0GsdGpGFSR8cL2DqYwZHQGAq+ueG6B76RaqkmxZmU+yZzxTUEqYzqKVM5o22sSbHPSgpSIvk5hzhIRR9vkgNKTZDuawe5NMIJwW+dC6hhr600Cot++G178W/edUPsxu56RpqVb5kQMgTgcQCfJ4TMgvVLbSrUReo+bxZPyIInYvCcHuFeDfg9Uq3UpHi6b2a11jCaWlGYS3lQOLFvMD1DS4OBgAprBPF1B8NyUKpbAeGEwmS0DoWikKbR8Rz8eUGU3F1ilPwcYMFtbDCDmWIUk2LHfmWpgtWa1TQ7iDEp0zu+OMyVkq1/Og1NVUApmxPG2Y4vnIgOLIe3D+HF33l/0+Mj4XSfw4UTrTAp1xopkO67ffdI6tcHEXQzKQnaJ/j5jmvoXSzXG8bdhDHgTa/WMOs2fu+IYVKaqkBTSHIz7xo3K7kEFSEQSPcFAqKmKlAVgppl46o7xobfIzldxW//YNemGPU84iTof04IeSfY+IwK2Mj4BwGcBfCh9Tm8zoH1STVegJqqYCjrDymjlKJmRTOpjECVxeGP6EgfTApZDYMZFddWalitNVrzDGb9nWLFsDuf7vOm8ybvN5arriXSIE/3Neb3w6PjObKa2nSzevLzhF3hjuEcZldrAV/ARscJAHh9inVChINUznUHKaf4bvPFemxaqRvgDMcK1Ddrpg1CWqsjcfD0X5kHqTaY1BaPSRlwHIqFsrG2dJ8ZClKBdB+rNcXXpJYrprcpEmEgsImbXq0hr6uRIgsO0Wh7jjTp1jTwa18phBOhz8ppTJDELaN4pqXfEferngZzm9gF4B8BfIFS+jfrclRdAGNSzRdg0JzTsB1QGn2h6iqJTfftKKSXnwNMjMFl6MGaFMB2WUF1n+jY1wIvZZaiV4q7TQSZFODO38mwQBQWTQAsP18O1YY8+XlC9/74cBbTkTUp9lnHeZAKpfsG3J+gWLM82XYU5kv1xEnGnQZnOIbtNNQicpraVpsBr6etGrwm5ar7WnCcGHU3SEsVA0sVA7ZDvbRcq8hnfDcTnu4LBhBNVbBlIBOZ7rNsB8Wa1dQPFsRAVkPVtOFQiplV1sibdO5YjVR8vdcsGxmXzawFaTz4RDUpgLVZ1C0bU8tVKCS+xtZPiOuT+lNK6b1gAw8XAfwVIeQEIeR3CCE3rtsRdgCUUreZt3khDXbr1yKk1BwipwCOiwuVhkGHabFzOIfplRqb+pn3b+TBgEKqW8IJID53zsFHvfMgFR4gFzaA5ciozcKJK26Q2pvApHa6/n2imhTfgZ6eLSKvq9gVupnzLpNKI56YLxktuyqsFR6TsoNMKjrNnIScrmIoq3lBiqekM1r6BVdTFYzkdSyVDcyVeCNve4tkEpMCWF0qKkituoFtNEb+PpBRQSlg2sk9UhxZTYm2RTLEnp2tIo0EPar+xZnU1EoNOwq5lpjw9Yw0tkgXKaX/iVJ6J5hL+UcAnEh4WU+hbjmgEBdFR/Ka18xbj9jhcGiqAsMS16TOL5Rjc+hR2DmSw+WlKqqm3cCkeGHYsh3ULafjwgnfviZZOLHk+fa5fVIhX75aDJMK90ldXapCV0liHWjHcA7zpbqXsgtuHPifHQoc3jHYVHfKBZhUHAzLwUrVXPcgxetQwQ0Pq+u1z5a3DWWwWneZlN06kwJck9mK2bbbBEdQ1CCqSbHjzWI+oibFWVjcvCk+QqVuc0uk5CCV09VIu6KaKbZDaxVJAg0g2gKLO6gz+blkURxp+qR0QsiHCCGfB+uPOgXgX3X9yDqISowv2oiASUXXpMTpvmKN3dg3bBsSvCoeO4dz3qIQlBbzPpCyZy7b+WZeIF1NyhvTMRjDpITCieY+qanlKnaN5BMFDePDWVDKeqqCnwkwYQl3+A7Xo4D0TIrLrbcV1rkmpTRbbLH+vPZ3zmNDWY9JtTpPioNZIxlrDlK5wKyyokDdB8T79y1zF5CYdF/eTavXbD/dl4QotSnQPN+pXbTSzBteZ7iN2LWVGnalMLPtF8QJJ95PCPkc2HDCj4O5lR+mlP4opfSr63WAnQC/KJJqUjWBUWoQumAOEACcmWWuB2EpdBoEi6MiJlUy+E2+cTUpf5aUmElFLbCiGsDVBPk5x7ibarrkzt0KB0G+oIiC1IAbpJJcJ+a9Hql1Tve5x9eY7osW7KTBtqFMIEi56r4W6yvcGokHqXZrUkzUkJTuy0YKJ3hLSFy6jzOppRpFzXTWzKRYuq9zTCquJlWus+nK4foXZ1Jp75F+QdxW67cAPA3gFkrphyiln6eUltfpuDoK3hSbF4gaGpkUn+kTJZxQmuYAAWsLUsGbK1iT4kHKm8q7zjUpPiIDYMKJjKZ4gaGZSYnTfSKD2atL1Vj5OQc/L5cWxEGKB0VRkOLrYZLrhF97WV8mpQmYVM1yGvwJW8XYUNZL97VjMAswprxUMTBfqiOnK23XQbl5MaUUpZrYHHlbIYOyYQuZPFeTxqX7+PU7XWHfNU5+zpFNYlIdSPclSd0BbjEmmJOlK5harsKwHKnsCyDyKqSU3r+eB9JNJKX7qqYNw3KEXedB6JqCarV5d35mroSMqmBfG812O6OYlNsH0o2pvIB4pEIQ/+7L38UXn7+CrYMZWLaDLQO6p54Km2HWTRtZQWooqzXWpAzLwUyxlqopkadvLi1WoCmkaTaPx6R2NItV0qb75ovtjaNYK/zZZI1MKkqwkwbbhrIomawe5c2TaqFPCvDHdfBG3nYNjfOuqKHu9tcNZZvNkX1rJAP7xxqv7eU0TMq9H2bK7LuOp0hNZnW1wUw6iFqH0n1AvNQdYDU3UQDOaqqXOUizkesXdH42dg+CB6modB/ArJF4KiAq3RdVkzozU8IN2wYjh5zFIRikGvqkMqwPpBuzpIBgM6/4Zjo5XcTBsQHcd2QbpldquCMwodbvqg/2SQmEE6Ga1MxqDZQCe1PcgGNDWagKwUrVFH73nK5CIcBBQfM0Z1KrCUyK+/atf5BqFk7UBYMXW8G2oQwoWDOu0S6TGsig5o4/WUuDc3ADFNUG4Df01rE/JDjiQp3wcEfRZ0yX2XfdmYJ55DQFsxFMqmY6ie0KaZHTlci5VYDr3iL4bjndt8vaPSKDFEdfBKm4qafDwSCVoO6LkqCfmSvh9j3tdedvG8wyw1GHNtyU+YzLpNyFdrCF/qs04Lv2KCf0xYqBuw5sxad+6E3NrxUxqQjHieCUUs++JsWCoipMATi9WhO/t65i39YB4W+lEIKhrJaY7psv1pHX1Y6z1CT4zbxBJ3kbuTaFCoAfaBdKhsekWg9S7Po7PVNqya8ujOAI+VJdvMnwmJRAPLFSMTCc02J7ljwmxdN9KeTyWV0VuvLzY+0UkxI1sQexVDFw087m3rxgylyq+3z0hRC/mpZJmcnCCdGIhcuLlSY/tbRQFOLVX4LpPl4Y5jdxp5kUISR2Oi93oRYhXByOFE6EzDa5mi6tmStP+YkC0X2Hx/D9t++KfO1QVkOpniycWG+3CcBPwwXbGdYunPAXfdN2oBC03JjK1ZvFutW2sg/wN4M8XR0WTQC+olIUpJarpncsUeD38kyFshlsKepJrA8p2mC2U/6N2QQmtVI1MZIX16QAdn/F+Rb2G/qCSSXVpAA2+FA0YTYITSVNO7Hz82U4tD3RBMf4cBZzxXrDQs9vQq606vSoDiB6Om/NtFE27MgbxbdF8tV9QuGE6gepwWxgBH2MtDgIVgxfETKpf/fBm2NfW8hpiX1SG9HIC4iZ1FqaeQF4wXahZMB0nLZSz8Hfe01BKiDDLtUsYf2FKyrni829UlHpsCD4/WE56Z0ZsrrYlZ8faz7GN7EV5DQ1UqBBKXUtn0TpPvaddseMselH9AmT4uq+pJpUfLpP5N13eg3KPo5dI3kUco3FZW6vxBVo3UhJBUcqBBG2QQrDk6An9Um555EHdl5rSLtL5EyqHdXbUE5LFk6U6hsbpAKsnJ3DDjEpi7blARj8vdcSpPi1WzFsFCOYVEZjDhdRTGokRtkX/AyghSAVMd8M6HC6LyYYFusWLIcK7y1+D8lUXyP6gkn5fVJiCTrAglQ9VZ9UY7rvzGwJCgFuaMMSieOn7z2Adx3d1vAYZ06zq3WoClmT8isKwZEKQXijuwfFu1lPgm7ZsB0K06Zix4lQP9VS2UA2IGVPAu+Vaue7D2XTMak794/GPqcb0BTfu49jrUxqOKdBI+w7WY7TsrIPCDGptQgnMv68r1LNijR+jbJGWqkYOLA13r2F3xN1y4mdyBtEeGIwB6W0o0Eqp6mRaUXeAyZqVOb31S4pmmhAXwSpuHTfsGDwYaxwIrRDOjtbwv6IAn5avOPQGN5xqLFQnQ8wqcFMe8ajSQiOVAjCb94V72a9NJ7pxAb28DTUxbKBLQPR01bD4Dvkdhbv4ZzumdmKYDsUi+WNYVLcLYMzqST3/TQghGA4SzDvjn1vx/dtJK+DEIBSfzBhO8jr/tDOON/JKNeJ5aqZKiU8kGHMKG4ibxBZTYUVEPJw1C3XWLpDJs45XcF8SbxBiru3guk+CR99ku6zoSniQrKuKhjMqJ5wQlWib3BdI03zkc4Ihu51Alw4Mbta77hogiNKOOEzKXGQ4rZENcsO1PEEBrOhtOBSJbkgHgTfIbeTBmPCiWgmtVQx4ND1ncjLwZkUTx0nue+nRSFDXOEEbdltAmD3B88srI1Jse9RqttszIwg3QewQBj273McipVqck0K8DMj6dN9Yl890RDCtYClFcVMajnkgyk6vt2ykbcBfRGkKoaNON0Bd52omU7k6HiguSZl2Q7OzZeaRkV0AvmAuq9bEulcRhX2SfEgFRdQuKM0v+HFfVLsXBq2m+6rGJEpRBHWwqSGcvESdL6DXwtjaBd+My87d0nu+2kxkiFMOGE7Lfv2cWx1d/idEE5w0U/UJmu7wBpptWaCUiTWpABfPJHGtw9onoPGkeTZ2SpYWlFckwpbjAWRlUxKiL4IUh++Yzd+/Oboi56P60hKueiqAsedqAowNwTTpm3Lz+PA+6Ish3YtSA3oKmqCdN9i2QAh8R3/OXf2Dd+FRvVJAQEmVRZ32keBp3HaZVJlw/Z+qzC4qmwjhRO8ncGbmdUhJmXZ1GNrrWLLYAaFnLamBTsfUqYWopjUUAbFutVQv0njNsEx4N4XaSyRgOYaKUenpvJycA8+EXwm1Xwf8NrdvoR6XL+hL2pSdx3citKF6IueM6mkmzO4A1YVdU2efUkI9nR1Ld2XUVERCCeWKgZG8nqsjLmJSQlHdTTuXJcqhrdTT4PRAR0ZVWmLSfGFsVS3hE4OC+WNsUQCfMcJLkH3pg+vlUllCRZmmONEu7OItg9lI62D0oIv9n6Pn/jeCyoS925hCzN3QE9Vk9I5k0oXpMKekxzc1aVjfVICz0oOHqRE1+QHb9+Jv8rfvSYR1vWIvghSSRjJ67i0WEF9KCsc3scRnqh6Zq6LQSoQmLrRIwWwm1IkQV8sJwcT5ihtxwon/D4pxmiWq+L+kCgQQvBDd+5py/0gKUh54yg2IEjx4M+l+UlOJ2kxnGE108Wy0XaQ+s0fuDlRFZmEjMZMVmeLzGEksiblBSnDD1It9NLx+2JHytQkz0iE3fF5Cq5Tjd1xTGqpYqCQ1YS/T05Xcf/NOzpyDNcTZJBCsCZlRzqgAwEm5S4uZ2ZK2Dmc65jnVxDB1EO30n35iJtpqWIkChzSMSm/UL1SZbWGVoQTAPCffvjNLT2fg+/eo+pS8yUDukoanOfXCz6TYum+TtWkhrPsfadX0s1XEuGAwAuxHeR1NbEmxeuBc4G61Eo1Oh0WxkBGQyGT3v4p2PAcBP97p0Q0bAy8A0ppk5J1pWrGzsmSaEZf1KSSMNJQk4pjUtwpgC0uZ+a6o+wDmNKKH0v30n2KN1IhiMWyeJRAENmwui9inhTAghTfra6X3QtnUlEzpeZLdYwNtu/0vRb4zbx+MzTQGSYFMCPfdhwnOol8RsVsQk2KizOCMvRWalI/ec8BfPTG9NfT9gi/QG+uWIdYddx03qWKkXhvSTRCBimwIMXNXONrUtxzje2SznZJfs7BxRNdE05kNNgObZLVL5briSq8nM5MNOOFE35NaqmcPCOok+AppmKEDH2+VF/3ibwcfjNvI5Nae5Bi/7ec9hwnOol8Ro2cysvBmUuQScXVbMJ4+w1b8T1707MSzqTCsvf5koGMqmA4Ipi2Ck9FKFD4LVfS9YBJ+NiQK5kQ8iOEkOOEEIcQctdGHEMQnH7PFuuxCwXv+zFtByeuFVE2bNy6a7hrx8VVUoMdajIMwyskB+pSlFIslU1sTZhW6zEpK3qB9fukbL/3ap2CFFdKRaX7FjbItw9gtTZdJR6TSjI2Tgue7gNanyXVaQTT1VE1qZyuYnw4iwsL/izV5Sqr2XSDCQ5kNAxkVCyEmNSCazTcKVbNf0eRyWzULCmJaGzUdus1AB8B8NgGfX4D+K5tNmTyGkZQOvzga9egEOB7b+leobPbTCo4UoGjbNgwbCc1k/Lk0zESdMN2fD/AFvqk1gJeJ4wSAWyUbx+Hpvg9d0mekWlR0An4OtuucKJTCPpkxo2ZObRtCOfm/CC1UuluzWZMYMW0UDY66obvZRAETGopwlxWIhobciVTSk9QSk9uxGeLwK2RbIemE07YDh48Po27D27t6kI34KqXuhakAh5rHDwtl7omFSecCPRJLSZYLXUaQ566r7kmRSnFQqmzC1Or0FXi90l1SDihKsQ7v3qPMKnBjBo7MuTQ9kGcmyt5ddGlitHVdBizYgqn++qeK3snEMWkbIditWZKJtUiel7dRwj5OICPA8D4+DgmJyfbep9SqRT52rPL/sW0OD8T+bwTc2xX/tXJYzg1Y+Anbsm0fTxpUC8z77mLZ05isnS2rfeI+97nZtj3efypZ3BxmC0q51bYubh6Lv4zF+frKJZtvHaC7TWef/ZpDGWaFyNNAU6fu+D9+dhTj3ddrFAqldjnAHjt5P/f3r0HuXVXBxz/Hj121/uy412vvXbs+JGn48ROYjsPaNg8OjWQIUlLCQXalKE82lBopzybzkA7wNDCQOnQwmR4lE5pTCYQCKFD4jw2JJQkzsNxgpM4D+dh7Ni79nrtfUur0z/uvZJ2I2klra6udHU+Mx57Ja30u766+un8fud3fi/Sn3ptxv2jCWcebvj1/fT3H/K1LfloKskrr+2nv3+A3a86HenjOx9mX0v5HdXIyAgt4vz+0cEBX9+bcxlzN7iMS6pgO1LDCY5PJPn5Xf10Nguvvj5OS4yi217o/Z3TxASvDOmM3/nd4Bgdi6MV+/96/rBzXf36oUc4sDDz5W1kSlGFwd+9Qn//gXm9RsnHXcd866RE5G5gWY67blTVnxX7PKp6E3ATwObNm7Wvr6+s9vT395Pvd1cOjPCFh+4HYO2qlfT1rc/5uPgLg/DYwzw/uRAY4KPXvNnXisU/fPVR9hw5xNbzN/KW05eU9RyFjlv2DsATj3D2xvO44JTFAOhzh+E3O7n0ogs4f9VJeZ/3vuGnefLIAVauXgvPPMsVfZfm3AplwX13snT5CkYnk3QdGeCyyy4r6zhK4R1z+/13snjpCvr6zp5x/wuHR+Ce+7lo03r6zlvhe3tyaf2/u+lZ2kNf37m8+OA+2LOHyy/9vXkNdfX393NKTwsHRo6woreXvr6NFWxxaW498Di7Bg7S1dma9/0Hzvvt5md3suz0jWxds5h/erSf1cs76es7v6jXKfT+zuXOo7vZsedw+ndUlZG7f8mGU1fR13dW0c9TSPT5AXj8ETZsPI8tqxenb983OAr39rNl41n0nXfyvF6j1OOuZ751Uqp6pV/PXWnZmUTFzEk9+MIgm1Yu8r2kvld1ot2nxbzpOamsxImjI8UlOHgLFr2hqqY8Q1Xe3jrFpLVXWkeeIrPpun0BzknFs+pApjMk55k4AZkMtqZYsMN96ffuHGsI13U72bEvDYywdc1ipwJ6EZl95epub+bo6CTTKSUaEUannCHrSg79ZipbzBzuS2/6mWNXXpOfpaAzu5OaOwV9OqVs25ArSKys1gASJ9Lrmea4aL3SLxPJaeJRyTvv0ByLMpVMcWws/3b0fslXZNZbvBlUCjq4nVTKq91XmTkpyHS8sUjAiRPx4r5grThpAU2xCC8NjpJKqZv95mPiRFsTKc1UtvAK3FZ0TiqWu/xSKdU0TEZQKejXish+4GLgFyJyZxDt8MSjkfQ3v2IiKYC3VqGTakunoFcvceLo6BSxiOTdqM7j1eU7MZEoWADW68yOBrCIsb05xokciRPePlPepopByE5Bn0w4OxtXYq7OWyAbfHaf8/6ZayF6NCKs7mrlpYERRqaSpNTf5JrujkwpJsjUcKxodl+60srMSGquHa9NboEkTqjqbcBtQbx2Pt6C3mLWSZ3V21mx8jGFZIb7/OqkMpvTebySSHN9YHrf+ofHkwUjgKZYhMnENEOjU1VLP/d0tMTT316z7TsyyqLWeMklmippRgp6Yn4bHmbzFsjWSnZfvuKy2dZ2t7P30InMrrU+Dvd5EZOzVqoj3VlVcug3XyQ1VGAvKZOfDfe5vAujUAq692G87Wz/oyiAdT3t9HQ05y0rM1+ZOamZkVQxC269SGp4PFHwA7Y55pReGh5PVG0hr6e9JZaz4sTLg6OsrsKXjELisUg6BX2+W8dn8z5sg4+knNcv5r27dkkbrx4dY8CdK/QzRXuJO8TrvdYRPzopLwV91pzU8JizBU6nD7U+w6zmU9CrxVsrVWjyetXiVr507TlctbG3Km26etMKrt7kX/ZZZk4q841vaDRRVMTjbStxfDxRMJJqjjmFRlNavZJIno7mWM7FvC8PjnLh2tIrq1dSPCIzFvOWs2dWLt6wVeAVJ4oc7gNYu6SdZEp5av8w4G+kkV153fnb6awqOV+aXQ4s29BYgoUL4kTK3OurUVkk5UpHUgWiAhHhPReuCs03Ia9zyZ6TOjI6WdQF60VSx8cTeTP7nMdFeN1dM1PtxImOlhjHxxMzCuiOT01zYHgi8D174tEIyazFvKGLpLzhviIjKYDHXx0CiisuW67OljixiKRLIx0ZmaSzJVbwPVyq5jyRlLNVjc1Hlco6KVcxnVTYRCLyhu06nLItc19I6UhqIlFwR9nmWCRrN9Lqdu6ru9uYTKbYPzSevu2Vo6Pp+4IUi0q6sO9cO0KXoqezme72Zk7pCnZ311LmU7009MdecTopP8siRSIyozTS4Gjlazg2Z9WszOZ35mJY2XCfKzMn1Vj99oKmKGNu4sS0mwJczL462XNS65YUTpzwVDuS8or/7jl4PL0l98uDTie1Jug5qWgkvTPvXPuYlaI5FmXnjVcEsgVJNi+SKmZOamFrnK62pvSXCT8TJ8BJnvDmoo74UMNRRHLuzjs0NhXIJpv1rrE+kQtoxEgKnN1NvQrlx8cTTgpwEZ2J15knpnWOSCpzX7WHOs5Y1oEI7DlwPH3bvsExAFZ3BxtpOCnomcSJSizk9QTdQUF2Bf/ivgd7Q36tTdGKzc/l093RnImkfKrhmGt33mNFjlKYmayTcjVqJ3Xx2i4e2DvIRGI6XQS2lDkpKLwINfu+aqd8tzbFWNPdxjMHszupEbrbm33ZTbkUsWgkPdw3mUz5/sFcbZtWLuLDl67l4nXFJaisdYf8/JyP8nS3NWXWSbnbdFRaSzySYzGv7cpbDuukXKf1tNMSj6QXQzaKt5+7nBOTSR54frDoCugwc9FzMZ1UUzTi275YhZxOTPh6AAAQiElEQVTV28merE7q5cEx1gQcRYHz/5FJnCi8I3Q9aolH+ezbziq6WooXSVUjA9SLpJLTKYbGEhWtNuFpjkVnLOZNTKcYmUxaJFWGcF0Z83DJqd3s/twfVH3eJGiXrOtiUWucX+w+wJHREiKprG/+xSyAPqktHsgw1PreTvYPjTM87iRv7DsS/BopcHbn9WMxb71au8SNpKoQaXS3NzGZTPHqUWfot9uHL6azI6lMtQmLpEplnVSWSqah1ot4NMK2s5exY88hDrrlgoqakyo6knI+fIP6Brl+uZM88ezB44xMJhk4MRl4Zh/MWsybrFwKer3KRFL+f4h7kdNzr58AnOG/SmuJR2fsJ+VVPllokVTJGvvKMAC8/dxeRqem+ekuZ4+boipOxKI5//3Gx7mRVFCdlJvh98zB4+nMvrW10EllRVJO7b7GjqRWLW4lFhEWVqFCuBc5Pet2Ul0+ZNw1xyIzduY9Nm6RVLksBd1w8douTmqNs+u1YyyIR3PuCzVbdvRUKDPNu6/adfs8PR3NdLU1sefg8fSHUS1EUrFoJF1g1iIpJ6L/0rXncPaKTt9fy1ti8Vy6k/InksreJqaU+V4zU2NfGQZwPjC3bXBKPRU7JzejkypUYDYabCQlIunkCS+SqoU5KWc/KSUxnWI6pRVbJ1XP3rVlJWcvX+j763jJUXsPucN9vkRS0ZlzUuP+F88NK+ukDABXnet0UsVGPLFohJhbg6xggVn3viATUtYv72TvoRGePzxC78KWoiJFv8WjQiKVSq+lafTEiWry3osvHxklHhU6fSjg7Gz2+cY5qSAr79cr66QMABeuWUxXW1NJ6bjeB2sxKejVLi6bbX1vJ1PJFPfvHaiJKAqcSEo1U4G+kot5TWHxaIRFrXFS6iRR+JF12hKLzpiTGhpLEItIIMsw6p3NSRnAiYy+9b4L0jXXitEcizAyOVfihBdJBTfMcZabPDE8nqiJ+SjIVCn3thKx4b7q6m5v5thYwpf5KPBS0LNqYo4Wt0+beSPrpEza1jWLS3p8MZFUU8DZfeCkNzfFIkwlUzWxkBcyc3XeViIWSVVXV1sTL+BPZh94i3kzkdTB4Ql6Fwa3E3Q9syvDlM3rnArNp5zW087JJy3gjGUd1WrWG8SjEU5f6iwWXeOW3wmaN5834nZSNidVXV4aeneVIqmDw+Ms67ROqhzWSZmyNRcRSa3ubuPBT19O78IF1WpWTt56qVqJpOLu/9nIpJP1Vej/0FSet4DXj8w+cL50JFOaXmZw8NgEyxcFew3UK7syTNm8D9Z6GKq6/MweTutpZ9Xi2piTikdmDvdZJFVdXudUzLY05UjvKZVMcWIiwYnJpA33lcnmpEzZvAWo9VAtYduG3vRasFrgJU54Cz6tk6oub7jPrzkp73xOJKbTNTF7LZIqS+1/BTY1y+ucGr1aQjni0dmRlP0fVlNXerjPvzkpcKqJHByeAGC5RVJlsSvDlC093FcHkVStic+KpOz/sLouWtfFey5cxebVpWW0Fss7n5OJ6XTh5mXWSZXFhvtM2YpJQTe5WSQVrM6WOF+69hzfnj8dSSVSHBieQASWWnZfWezKMGWrp8SJWhOLetl9tpg3jLzM14mkE0n1dDSnv5iY0tj/milbJpKyD9hSecN9JyacFHRLnAiXdHZfIuUu5LWkiXJZJ2XKllnMa2+jUnnfqr3FvDZkGi4tWZHUgeFxli+yob5y2ZVhyuZdiE02jFGydMWJySRN0QiRiNV0C5OWrMSJ14cnWNZpkVS5LHHClO2a85ZzUltTen7FFC87ccLm9MLHO6eHT0wyNjVtkdQ8WCdlynZqTwen9gRXk6+exbMSJ2w+Kny8c7rP3WjT5qTKZ1/hjAlA9jopm9MLnxZ3jjHdSVkkVbZArg4R+YqIPCsiu0XkNhFZFEQ7jAmKF0lNp9SyI0OoeVYktdwiqbIF9RVuB7BBVc8F9gKfDagdxgQie82MRVLh40VS+4fGiUaEJR3+1AhsBIFcHap6l6om3R8fAk4Ooh3GBMUrMAu2kDeMYtEIsYgwnVKWdjQTtezNsomqBtsAkZ8DP1LV/85z/4eADwEsXbr0gu3bt5f1OiMjI7S318aGd9XUiMddD8c8mlBuuGcMgLO7Inxyy/yHg+rhuCutlo/5IztGmZiGUxdF+IeLKjvcV+pxX3bZZY+p6uaKNqJKfMvuE5G7gWU57rpRVX/mPuZGIAn8MN/zqOpNwE0Amzdv1r6+vrLa09/fT7m/W88a8bjr4ZjHppJwz50ALFvSTV/flnk/Zz0cd6XV8jG3PbCDidEpzly1lL6+8yv63LV83JXmWyelqlcWul9ErgeuAq7QoMM5Y6osFsmek7LhvjDyzqvtyDs/gayTEpFtwKeBt6jqWBBtMCZI8aw5KVvMG05eqatlVv18XoK6Or4JdAA7RGSXiHw7oHYYEwgRSZdGskgqnJrTkZR1UvMRSCSlqqcG8brG1JJ4NEIyNW3ZfSHlLS2wahPzY+MMxgTES0O34b5w8ob7rNrE/NjVYUxAvOrxFkmFU0s8SjwqdLfZQt75sE7KmIB4kZRVnAintqYYvQsX2DYs82RV0I0JiJeGbokT4fSxK05jeDwRdDPqnnVSxgSkyZ2zsF15w+mMZbaNTSXY1WFMQCwF3Zi5WSdlTEC8Sug2J2VMfnZ1GBOQeDoF3SIpY/KxTsqYgHiRlM1JGZOfXR3GBCSTgm6RlDH5WCdlTEDitpjXmDlZJ2VMQCxxwpi52dVhTEC8FHRLnDAmP+ukjAlIPOYN99llaEw+dnUYE5C4LeY1Zk7WSRkTkMyclHVSxuRjnZQxAYlFI8SjQtSqZBuTl3VSxgSkORaxKMqYOVgVdGMCcv0lq7l4XVfQzTCmplknZUxA1nS3saa7LehmGFPTbLjPGGNMzbJOyhhjTM2yTsoYY0zNsk7KGGNMzbJOyhhjTM2yTsoYY0zNsk7KGGNMzbJOyhhjTM0SVQ26DUUTkQHglTJ/vRsYrGBz6kUjHncjHjM05nE34jFD6cd9iqou8asxfqqrTmo+RORRVd0cdDuqrRGPuxGPGRrzuBvxmKGxjtuG+4wxxtQs66SMMcbUrEbqpG4KugEBacTjbsRjhsY87kY8Zmig426YOSljjDH1p5EiKWOMMXXGOiljjDE1qyE6KRHZJiLPicgLIvKZoNvjBxFZKSL3icgzIvJbEfm4e/tiEdkhIs+7f58UdFsrTUSiIvKEiNzh/rxGRB52j/lHItIUdBsrTUQWicitIvKse84vDvu5FpG/dd/bT4vIzSLSEsZzLSLfE5HDIvJ01m05z604/s39bNstIucH13J/hL6TEpEo8O/AW4H1wJ+IyPpgW+WLJPB3qnoWcBFwg3ucnwHuUdXTgHvcn8Pm48AzWT//M/B195iHgA8E0ip/fQP4paqeCWzEOf7QnmsRWQF8DNisqhuAKPBuwnmu/xPYNuu2fOf2rcBp7p8PAd+qUhurJvSdFLAVeEFVX1LVKWA7cHXAbao4VT2oqo+7/z6B86G1AudYf+A+7AfANcG00B8icjLwduA77s8CXA7c6j4kjMfcCVwKfBdAVadU9RghP9dADFggIjGgFThICM+1qv4KODrr5nzn9mrgv9TxELBIRHqr09LqaIROagXwWtbP+93bQktEVgPnAQ8DS1X1IDgdGdATXMt88a/Ap4CU+3MXcExVk+7PYTzfa4EB4PvuMOd3RKSNEJ9rVf0d8FXgVZzOaRh4jPCfa0++cxv6z7dG6KQkx22hzbsXkXbgx8DfqOrxoNvjJxG5Cjisqo9l35zjoWE73zHgfOBbqnoeMEqIhvZycedgrgbWAMuBNpyhrtnCdq7nEvr3eyN0UvuBlVk/nwwcCKgtvhKROE4H9UNV/Yl78yEv/Hf/PhxU+3zwJuAdIvIyzjDu5TiR1SJ3SAjCeb73A/tV9WH351txOq0wn+srgX2qOqCqCeAnwCWE/1x78p3b0H++NUIntRM4zc0CasKZbL094DZVnDsX813gGVX9WtZdtwPXu/++HvhZtdvmF1X9rKqerKqrcc7rvar6XuA+4J3uw0J1zACq+jrwmoic4d50BbCHEJ9rnGG+i0Sk1X2ve8cc6nOdJd+5vR34MzfL7yJg2BsWDIuGqDghIm/D+YYdBb6nql8MuEkVJyJvBh4AniIzP/P3OPNStwCrcC70P1bV2ZOydU9E+oBPqOpVIrIWJ7JaDDwBvE9VJ4NsX6WJyCacZJEm4CXg/ThfOkN7rkXkH4HrcDJZnwD+Amf+JVTnWkRuBvpwtuM4BHwO+Ck5zq3bYX8TJxtwDHi/qj4aRLv90hCdlDHGmPrUCMN9xhhj6pR1UsYYY2qWdVLGGGNqlnVSxhhjapZ1UsYYY2pWbO6HGNN4RKQLp5AnwDJgGqcUEcCYql4SSMOMaTCWgm7MHETk88CIqn416LYY02hsuM+YEonIiPt3n4jcLyK3iMheEfmyiLxXRB4RkadEZJ37uCUi8mMR2en+eVOwR2DCRESWich2EXlRRPaIyP+KyOkVfP5pEdnl7uP1cxFZ5N6+XERuLeL3R/Lcfk0x2yZZJ2XM/GzE2c/qHOBPgdNVdStONYi/dh/zDZw9j7YAf+TeZ8y8uRUnbgP6VXWdqq7HqTSztIIvM66qm9x9vI4CNwCo6gFVfWfhXy3oGpw9/gqyTsqY+dnp7uU1CbwI3OXe/hSw2v33lcA3RWQXTq21ThHpqHpLTRhdBiRU9dveDaq6C3hQRL7iRj9Pich1kI7+fyUit7lR17dFJCIiHxCRr3vPISIfFJGvvfHl+A3uViAistrbPditqXiLuzvwj9zdkjdnPd8XReRJEXlIRJaKyCXAO4CvuFHaunwHaIkTxsxPdp24VNbPKTLXVwS4WFXHq9kw0xA24OyrNdsfAptwIv1uYKeI/Mq9bytOBPMK8Ev3sduB3SLyKbfK/PuBD2c/obvL+RW4m23O8lfAkKqeKyIbgF1Z97UBD6nqjSLyL8AHVfULInI7cIeqFhwytEjKGP/dBXzU+8EtDmuMn94M3Kyq06p6CLgf2OLe94i7U/k0cDPwZlUdBe4FrhKRM4G4qj7lPn6BOwpwBKeQ7448r7cdQFWfBnZn3TcF3OH++zEyIwxFsU7KGP99DNjsDoXsAT4SdINMaPwWuCDH7bk2Q/TMTun2fv4O8Oc4UdT3s+4fV9VNwCk4VfdvKPH1EppJI5+mxBE866SMmYOqfj47/VxV292/+1X1qqzb+7xtErLvU9VBVb1OVc9V1fWqap2UqZR7gWYR+aB3g4hsAYaA60QkKiJLgEuBR9yHbHX314vgbH3yIIC7ieZK4D04EdYMqjqM84XrE+4Gq9keBN7lvv56nESiuZwA5pybtU7KGGPqlBuhXAv8vpuC/lvg88D/4Ay5PYnTkX3K3SwTnOSHLwNPA/twsgM9twC/VtWhPK/3hPuc7551138AS0RkN/Bp97WH52j+duCTIvJEocQJW8xrjDENIntz0Dz334GzXOKeXPcXeN4ozjzWhNvh3IOzHGNqvm227D5jjGlw7gLdR4AnS+2gXK3Afe4woAB/WYkOCiySMsYYU8NsTsoYY0zNsk7KGGNMzbJOyhhjTM2yTsoYY0zNsk7KGGNMzfp/8LFdNlpe3SwAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(wn)\n",
    "plt.title('White Noise')\n",
    "plt.xlabel('Time')\n",
    "plt.ylabel('Value')\n",
    "plt.figtext(0.995, 0.01, 'CopyRight', ha='right', va='bottom')\n",
    "plt.grid('True')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import data from yahoo finance\n",
    "data = ffn.get('^gspc',start='2020-01-01')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Date\n",
       "2019-12-31           NaN\n",
       "2020-01-02   -100.872724\n",
       "2020-01-03    -73.802655\n",
       "2020-01-06    -96.802655\n",
       "2020-01-07    -85.372724\n",
       "                 ...    \n",
       "2021-03-17    631.057208\n",
       "2021-03-18    642.467364\n",
       "2021-03-19    583.807208\n",
       "2021-03-22    581.447345\n",
       "2021-03-23    608.937335\n",
       "Name: gspc, Length: 309, dtype: float64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(data.gspc.shift()-data.gspc.mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Date\n",
       "2019-12-31   -100.872724\n",
       "2020-01-02    -73.802655\n",
       "2020-01-03    -96.802655\n",
       "2020-01-06    -85.372724\n",
       "2020-01-07    -94.472821\n",
       "                 ...    \n",
       "2021-03-17    642.467364\n",
       "2021-03-18    583.807208\n",
       "2021-03-19    581.447345\n",
       "2021-03-22    608.937335\n",
       "2021-03-23    578.867267\n",
       "Name: gspc, Length: 309, dtype: float64"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(data.gspc-data.gspc.mean())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "44397904.98424241"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.sum((data.gspc-data.gspc.mean())**2)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9883216684411759\n"
     ]
    }
   ],
   "source": [
    "# 一階自我相關係數\n",
    "# cov(y_{t-1},y_t)/var(y_t)\n",
    "cov = np.sum((data.gspc-data.gspc.mean())*(data.gspc.shift()-data.gspc.mean()))\n",
    "var_1 = np.sum((data.gspc-data.gspc.mean())**2)\n",
    "var_2 = np.sum((data.gspc.shift()-data.gspc.shift().mean())**2)\n",
    "ar_1 = cov/(np.sqrt(var_1)*np.sqrt(var_2))\n",
    "print(ar_1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "136964.22547289604\n",
      "136964.225472896\n"
     ]
    }
   ],
   "source": [
    "print(data.gspc.var())\n",
    "v_1 = np.sum((data.gspc - data.gspc.mean())**2)/(len(data.gspc)-1)\n",
    "print(v_1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9875562625490979"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# correlation \n",
    "data.gspc.corr(data.gspc.shift())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>twdx</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2018-12-31</th>\n",
       "      <td>30.552999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-01</th>\n",
       "      <td>30.565001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-02</th>\n",
       "      <td>30.500999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-03</th>\n",
       "      <td>30.759001</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-04</th>\n",
       "      <td>30.863001</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 twdx\n",
       "Date                 \n",
       "2018-12-31  30.552999\n",
       "2019-01-01  30.565001\n",
       "2019-01-02  30.500999\n",
       "2019-01-03  30.759001\n",
       "2019-01-04  30.863001"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# import exchange rate (TWD/USD)\n",
    "twdusd = ffn.get('TWD=X',start='2019-01-01')\n",
    "twdusd.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.9957863613822104"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# calculate the first-order autocorrelation\n",
    "twdusd.twdx.corr(twdusd.twdx.shift())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "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>twdx</th>\n",
       "      <th>t_trend</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2018-12-31</th>\n",
       "      <td>30.552999</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-01</th>\n",
       "      <td>30.565001</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-02</th>\n",
       "      <td>30.500999</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-03</th>\n",
       "      <td>30.759001</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2019-01-04</th>\n",
       "      <td>30.863001</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 twdx  t_trend\n",
       "Date                          \n",
       "2018-12-31  30.552999        0\n",
       "2019-01-01  30.565001        1\n",
       "2019-01-02  30.500999        2\n",
       "2019-01-03  30.759001        3\n",
       "2019-01-04  30.863001        4"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# construct the time trend variable\n",
    "twdusd['t_trend'] = range(len(twdusd.twdx))\n",
    "twdusd.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x261f9d48278>"
      ]
     },
     "execution_count": 89,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYYAAAEFCAYAAADnpECTAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjAsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+17YcXAAAgAElEQVR4nOydd3gc1dW436Ot6l22bLkX3HABYZsOjgFTAk5CL4EE4jSSfOnkS/uA1F8KCQkplCSEEiBAgDiUmGKqO+7dlpuKrd777v39MbOrlbSypVVd+bzPs49m78yde2e1O2dOueeIMQZFURRFCRAz2BNQFEVRhhYqGBRFUZR2qGBQFEVR2qGCQVEURWmHCgZFURSlHSoYFEVRlHY4I+kkIl7gHcBjn+NZY8wPReRO4H+ASUCmMaY0TN8LgftCmqYB1xtjXhCRvwHnA1X2vtuMMZuON5eMjAwzfvz4SC5DURTlpGXDhg2lxpjMcPskknUMIiJAvDGmVkRcwHvAV4AmoAJYCeSGEwwdzpMG7ANyjDH1tmBYbox5trtzyc3NNevXr+/xNSiKopzMiMgGY0xuuH0RaQzGkia19luX/TLGmI32gN091dXAK8aY+kjmoSiKovQ9EfsYRMQhIpuAYmCFMWZNBKe5HvhHh7Yfi8gWEblPRDyRzk9RFEWJjIgFgzHGZ4yZC+QA80VkVk/6i0g2cCrwWkjzd7B8DmcAacC3u+i7TETWi8j6kpKSiOavKIqihKfXUUnGmEosn8KSHna9FviXMaYl5FxFxqIJ+Cswv4sxHzTG5BpjcjMzw/pOFEVRlAiJSDCISKaIpNjbscBiYFcPT3MDHcxIthYRcG4vBbZFMj9FURQlciLVGLKBt0RkC7AOy8ewXES+LCL5WOalLSLyMICI5Aa27ffjgTHA2x3O+4SIbAW2AhnAjyKcn6IoihIhEYWrDiU0XFVROrOzqJoEj5MxaXGDPRVliHK8cFVd+awow5CvPbOZn7/aU+uuolioYFCUYUh1Qws1ja2DPQ0lSlHBoCjDkMYWH02tvsGehhKlqGBQlGFIQ4uPxhb/YE9DiVJUMCjKMMMYYwsG1RiUyFDBoCjDjGafH2OgqVU1BiUyVDAoyjCjsdkSCKoxKJGigkEZthRWNvDj/+ygxXdyPTk32AJBBYMSKSoYlGHLz17ZxUPvHuDt3SdXosXGoGA4uQSi0neoYFCGLTF2WZAdRdWDO5EBJqgxtPqI9swGyuCggkEZtlQ2WIl71x0sH+SZDCwBwWCM5YhWlJ6igkEZthwutwoDfnio4qRa7NXY3HatGpmkRIIKBmVY4vMb8ssbmJ2TTF2zj6fXHRnsKQ0YjSFCUB3QSiSoYFCGJceqG2n2+bnujDHMH5/GA2/tO2lukg3NbVpCkzqglQhQwaAMSwJmpHFp8fzPRVM4Vt3EP9YeHuRZDQwNLaoxKL1DBYMyLCmqagBgVIqXsyZlsHBiGn9Yuf+kuFG2FwyqMSg9J9LSnl4RWSsim0Vku4jcbbffKSL7RMSISMZx+vtEZJP9eimkfYKIrBGRvSLytIi4I5mfohRXNwGQleQF4KuLp1JS08Tjqw8N5rQGhKZQwXASOd2VviNSjaEJWGSMmQPMBZaIyELgfaz6zyf69TUYY+barytD2n8O3GeMmQJUALdHOD/lJGP5lkK25FcG3xfXNBHndpDgcQKwYGI6p41N4YVNBYM1xQGjoVlNSUrviEgwGIta+63LfhljzEZjzMFIzikiAiwCnrWbHgWWRnIu5eTCGMNdz23lM39fT3WjtXbhWHUjWYmedseNz4inoq5lMKY4oKgpSektEfsYRMQhIpuAYmCFMWZND7p7RWS9iKwWkcDNPx2oNMYEyk7lA6MjnZ9y8lBc00RtUyvHqpv46cu7gm0BM1KAJK8rKDi6ywf7Stl1NLpWTocKA9UYlEiIWDAYY3zGmLlADjBfRGb1oPtYuwj1jcBvRGQSIOGGCddZRJbZgmV9ScnJlQdH6cz+Ekt5nTsmhX+sPcyq/WWU1DR10hiSY13UNLbi83cvTcSD7+znxofXcPvf1kfVDTZUY9AFbkok9DoqyRhTCawElvSgT6H9N8/uOw8oBVJExGkflgMUdtH/QWNMrjEmNzMzM/LJK8OCvJI6AH517RzGpsXxnee3cLi8nqzEDhpDrAuA2hPUQvb7Dfcu38FPXt5F7rhUCiobeHJN9IS6Nrb48LpigtuK0lMijUrKFJEUezsWy+G8q5t9U0XEY29nAGcDO4yV7est4Gr70FuBFyOZn3JykVdSR6zLwYT0eH728VM5WFaPz2/ISmqvMSR5rWeOE5mT/vTOfh557wC3nTWeZz57JtnJXrYXRo85qaHZR0qsFdCngkGJhEg1hmzgLRHZAqzD8jEsF5Evi0g+1tP+FhF5GEBEcgPbwHRgvYhsxhIEPzPG7LD3fRv4mojsw/I5PBLh/JSTiLzSWiZkxBMTI5w1OYNJmfEAwYikAAGNoarh+ILhzZ3FzBmTwg8/OoOYGCHe46S++fhaxlCisdVHSpx1rWpKUiLBeeJDOmOM2YJl/unYfj9wf5j29cAd9vYHwKldnDcPmB/JnJSTE2MM2wurOXtSerDtvuvm8sm/rGXBhLR2xybbguF4GkNTq48tBVXceuY4rEA5iHM7qG+Onifv6oYWUuJciKjGoESGrnxWopr9JXWU1DSxcGKbYJidk8KmH1zMlBGJ7Y5N8tqC4Tgaw/bCappb/Zw+LjXYFud2tFsbMJTx+w27j9YwdUQiHmeMCgYlIlQwKFHNqv2lAJwZojF0RVKs7WNo6Nos9OGhCgBOGxsqGJzUt0SHKelgWR11zT5mjUomzu3kw8OV7TSkFzYWMP6u/5zQnBbAGBM1QlHpO1QwKFHNqrwyRiV7GZsWd8Jjk7phStpXXEt6vLvdGojYKDIlBZzkM0cn8ZWPTGHDoQoeW9WWiOBnr1gxIqW1Td0633MfFjD/x6/3eP2HEt2oYFCiFr/fsDqvnIWT0oP+gOOR4HYSI8c3JR0orWN8Rny7tjiXg/qm6BAM2wqrcDmEKVmJ3HrWeBK9znZC4Gh1I0C3S36+uesYNU2tbMuv4v19pby0OWwEuTLMiMj5rChDgT3FNZTXNXPmxBObkQBiYoREr4vq46xjOFhWxzmT26+NiaaopO0F1ZwyMhG303rmi3c7qWvqPPfuRCsZY1h7wDKtbSmoCmobV84Z1YczVoYiqjEoUcuq/WVA9/wLAZJinWHt67uP1nDV79/jWHUTEzPbawyxbke71cRDFWMM2wqrmDUqOdgW73FQZ2s7/pAV383dEAwHy+qD2kZogkJl+KOCQYlKWnx+nt2Qz7j0OHJST+xfCJDkdYUVDM9uOMLm/CoAxqd3NiW1+Ey3bqaDSUFlA5X1Lcwc3SYYEjxO6mxtp8g2I0H3BMO6A+UATBuZyOYjVcF2fzdTiijRiwoGJSp5fPUhthdWc9eSaT3qNyY1joOldZ3a39rdlnNrfEZ7QRNnL5Qb6tE5QcfzqKRgW1yIKSmvpDbY3uw7sWBYe7CclDgXS+eNpqCyIdheE2Ka2nSkktowpiolulHBoEQlq/PKmJgRz6WnZveo3/TsJA6U1bXzGRwpr2dfcS3nT81kzpgUJmUmtOsT53YADPmQ1e0FVcQITB/ZJhjiPU5qbVPStoK2tB53PbeV8Xf9p13/w2X17fwR6w+WkzsujTk5Ke2Oq6q3NK59xTUsfeB9Lv7120NeaCo9QwWDEpXsL6ljclbCiQ/swIxRSRgDu47WBNte234UgLuvnMmLXzwbr8vRrk9QMAzxm9+2wmomZyUQ626bf4LHEbzZbz7S5icIaACttubg8xuu+N27/HrFHgCKaxo5WFbP/AmpzBrdJmgAKhuaAfjPFutzK6xq5M1dxf10VcpgoIJBiTpafX4OldUxMbPngmF6trUaekdIUryXtxYxIzupU5hqgFhbUAz1p+JtBe0dz9A+omrTkcpON/lAhFZeSS3Vja28v89aMLjOjkY6Y3waiV5XO4d8wEfz8tYixqVbZreAsFCGByoYlKjjSEUDLT4TTJbXE0anxJLkdbK90HKmFlY28OHhSi6f3bVJKt72MQxljaG4ppHimqZ2jmcImJJaOVrVyNHqRuaPbx/BVVlv3dB3FFmCctfRGirqmll3sJxYl4NZ9vnmjkkJ6dPChkPl7D5Www3zxwKEDYlVohcVDErUEXCiRqIxiAhnTkrnzV3F+P2GV7ZZ5pBLZ43ssk/ANFM3hNcyhHM8g7WOobHFz4eHLQ1gfofEghW2vyA0rfi6g+WsPVDOvLEpuBzWLeKuJdP4y225AFQ2tPDQOwdIjnVx88JxAEE/hjI8UMGgRB2Bim2RaAwAl52azbHqJj48XMErW4uYNjLxuEIm4GMYyqakgGlsRkfB4LHm/sH+UpwxwmljOziSbRPQ9sIqThmRSKzLwctbi9h5tJozxrcJkawkL2dPzgBgy5FKXttxlJsWjCXB4yTe7VCNYZihgkGJOvJK6kiPd5MS546o/0emj8DtjOGv7x9k/aEKLj9BZFOca+ibkspqm4l3O4IZZAMEalJ8sK+MadmJJHbYX1nfQl1TKxsPV3LGhFTOm5rBC5sKMQbOnZLR7liP00Gc28HzGwtwxgi3njUesMxVKhiGFyoYlKhjf0ltp5DSnpDgcXL+1Ez+s7UIgMuO418AiPMENIahe/OraWzpdNOHtjUYeaV1zMlJCabKCFBR38LLW4uob/axdO5oLpphmdRm5yS3Sz0eIDnWhc9vuHLOaEbYiQYTbD+GMnyItLSnV0TWishmEdkuInfb7XeKyD4RMXbZznB954rIKrvfFhG5LmTf30TkgIhssl9zI7ssZTizv6SuU9qKnnLZqdYN8JQRiScUMnFBH8PQ1RhqGltJ9HZOfZbgaQtdnTsmBUeM4IhpSzhYVd/MP9fnMzEjntPHpbJ4ehaTMuP5+sWnhE1MGCh2dPs5E4JtqjEMPyJNotcELDLG1IqIC3hPRF4B3geWAyuP07ce+KQxZq+IjAI2iMhrxphAkPU3jTHPRjgvZZhTUddMeV1zrzQGgMXTR5DkdfLx00af8Fivc+ivY6htCi8Y4t1tbYHIIrcjhga/dS0bj1Sy9mA537zEEgQpcW7e+PoFXY4zc1Qy49Pj2/ky4txt+ZiU4UGkpT0NEFhf77JfxhizEThuCmRjzJ6Q7UIRKQYyAc3SpZyQvNJARFLvNIZEr4v37lpEgvvEP4GYGLHMJcfJyjrY1DS2kBzG5xIItU3wOIMOdrczJpgU8N29pcQIfOK0nG6N86tr53RK2Z3gcVJU1dhFDyUaidjHICIOEdkEFAMrjDFrIjjHfMAN7A9p/rFtYrpPRDxd9FsmIutFZH1JSUm4Q5Rhyv4SK89RbzUGsBLqxcScuI4DQKLXSc0QLlbTlSkpIBhm5yQHTUgd/QznTc1kZLK3U9+u6PjgFx+SqE8ZHkQsGIwxPmPMXCAHmC8is3rSX0SygceATxljAhm9vgNMA84A0oBvdzH2g8aYXGNMbmZmZrhDlGHK/pJa3I4YclJjB3RcSzAM3ZtfdWMrSWF9DFbbnJAFap4OguHa3DG9Glt9DMOPXkcl2b6BlcCS7vYRkSTgP8D3jDGrQ85VZCyagL8C83s7P2V4kVdSx7j0OJyOgQ2oS/K6hnR5y66ikjIS3HxrySnctGBssC1UY0iLd/OR6Vm9GjvBoz6G4UakUUmZIpJib8cCi4Fd3ezrBv4F/N0Y888O+7LtvwIsBbZFMj9l+NLbUNVIGcoaQ3Orn6ZWP4mezhqDiPCFCya3q1nhtoXqZ8+byL++cBYep6NTv54Q73HS0OLDp3Uahg2RPnZlA2+JyBZgHZaPYbmIfFlE8rHMS1tE5GEAEckNbAPXAucBt4UJS31CRLYCW4EM4EcRzk8ZhrT4/Bwuq++14zkSEr2uIetjCKwhCOdjCEfAlDQy2cu49N5/lgFzlfoZhg+RRiVtAeaFab8fuD9M+3rgDnv7ceDxLs67KJL5KCcHh8vrafWbQdMYjlcrejAJCKyEMKakcARMSR3Ti0dKwMFd19TaaeW1Ep3oymclasizI5IGQ2NIirU0ho6hmkOBgImruxpDm2Dom59/qGBo9fl5Y+cxnv8wP1jrQYk+Il3gpigDTnGNFSufnTywEUlg3XRbfIamVn/ET9q1Ta0cLK0LprLuKwJO8W4LBtvH4O2lbyFAYHX1Pct3sudoDUft2tL/XJ/PY7fPH/BAAaX36H9MiRrKaq1MoGnxkSXP6w2BiJ/qhsj9DL97cy9X/O49nlxzuK+mBRBceNddM05fm5KyEq01EO/sKeGUkYn86ebTuOeqmazKK+Ph9w70yRjKwKIagxI1lNU2keR1dlqgNRAE1ghUN7aSlXSCg7tga75VHOjuf29n6bxRxHVj1XV36KkpKRCF5OkjU9LMUUks/9I5ZCd7SU+w1qQaY3h/Xym/XrGHxdOzmJyV2CdjKQODagxKVLDhUDnFNU1kJIRdDN/vBJ7GexOZdKC0jsxED02tfj7YV9ZXUwvOKdw6hnD0tcYgIswanRwUCoG2e5fOIs7t4Ov/3KL+hihDBYMy5Hl7Twmf+OMqXtl2lPSEgTcjQdvTeKSRSVX1LRRVNfLJheNI8Dh5Y1dxn80tUIUt3MrncAQFQx/5GLoiK9HL3VfOZPORSh56V01K0YQKBmXIszqv7ek6PX5wNIbEXmoMu45aFdZOzUnmzEnprDnQdxpDZX0zSV5nt528QedzH5mSjseVc0ZxycwR3LdiD3uP1fT7eErfoIJBGfKsPVAe3B4sjSEp1noaj3T1866j1k1xenYSOamxHOvDbKQV9S09csh7+tiUdDxEhB8tPZV4j4NvPKsmpWhBBYMypKlramXTkbaM7OmD5GMIrO7tjcaQEuciK9HDiCQvdc2+Pks8V1Hf3KMyp33tYzgRmYke7r5qlpqUoggVDMqQpqy2uV0OnrS4wVlZG2vfRBtbInvi3XW0hmkjExERshIt4VZc09Sjczyx5hDv7yvt1F5R30xqDz4XTx8vcOsOH52drSalKEIFgzKkqWmyntADN9PWQUrU5nTEWJXPWnqeRdTvN+w+WsO0kVacayDu/1h1z8xJv3htN3/74CBgaVI/fHEbpbVNVNS1kNoDjeHKOaO5+8qZfRYu2x3amZT+uVlNSkMcFQzKkCZg0//onFEAwZvrYOB1xdAQQXnP/IoG6pt9TBtpxfJnJfVcY2ho9lFZ30JhZQMAz32Yz6OrDrFixzEqe2hKGpsex61nje/+BfQRQZNSfpWalIY4KhiUIU1gVe9Vc0ex9n8/wjlTMgZtLrFuR48FQ4vPz047ImlatiXURtgaQ3EPNIbCKksgFFU1YowJrp7eXlhFXbOvR6akweSjs7NZMnOkmpSGOCoYlCFNIKV0gsdJVlL3y0/2B7EuR49MSQdK65jy3Vf4zet7EYGpI6yssEmx1urtgMbg8xvuXb7juDfKo3YUU3ldM6vyyth1tAaRtoit1EFIExIJgYVvalIa2qhgUIY0NQHB0M3FW/2Jt4eCYbcdorqzqJpxaXFBm37AAR3QGNYfLOeR9w6wfEtRp3MYY/jeC1u549H1wbb7Vuwhzu1gycyR7DlWC9AjH8NgE2pSevDdvMGejhIGFQzKkKanCeL6k1i3g8YeCIaqhubgdkffyIgkL8eqLY3h9Z3HADhSXt/pHE+vO8Ljqw+3E0jrDlZw5ZxRzBvbVsc5WkxJAQImpd+s2KsmpSFIxIJBRLwislZENovIdhG5226/U0T2iYgRkS4NwiJyq4jstV+3hrSfLiJb7XPcb5f5VE5SaptacMZIpwL2g0Gsq2c+hqNVbc7lU0a2TyI3KiWWwqoGjDGs2GEJhsNhBMMLmwrCnvuG+WOZndMmGHrifB4KqElpaNObX1sTsMgYMweYCywRkYXA+1g1oA911VFE0oAfAguA+cAPRSTV3v1HYBkwxX4t6cUclSintrGVBK+TofB80FMfw9EQ5/L07PaCISc1lsLKBvYW13KwrJ5Yl6OTYPD5DVvzq7jm9Jxgm4i1enp2TjILJ6bzwI2nsey8iUH/RTSRmejhHjUpDUkiNtwaq5RVrf3WZb+MMWYjcKIf8iVYdaLL7WNXYAmWlUCSMWaV3f53YCnwSqTzVKKPx1YfwiHCjQvGUtPUGlx1PNh43T0TDMeqGxmbFsclM0dw7pTMdvtyUmNp8RmeWG09P338tNE8seYwl/72XaZnJ/LjpadypKKeumYfCyemU9/sIyctFkE4c1J68Pd1+exsLp+d3XcXOcBcMTubl7cW8ZsVe1k8fQRTR2h67qFAr35xIuIANgCTgQeMMWu62XU0cCTkfb7dNtre7tjecdxlWFoFY8eO7fnElSHNg+/sp6y2mSvmZFsawxARDHEuB409MCUdq25kclYC3718Rqd9OalxADy7IZ9Zo5OYPyGNJ9YcZmdRNTuLqrlh/lgOlFqlTOeMSeETIVrDcEJEuOeqWazOe5tv/HMzz3/+LK34NgTo1X/AGOMzxswFcoD5IjKrm13DqRPmOO0dx33QGJNrjMnNzMwM00WJVhqafcEFYc+uz6e2qbXbBWj6m9gINIYRSeFzO+WkWuVJ65p9LJ4+Ivg+QGlNE1vyK0n0OJmYMfA1rgeSgElpi5qUhgx9IpqNMZXASrrvD8gHxoS8zwEK7facMO3KScL+klqMsRK9PbrqIFUNLUNGY4h1OajvoDHsL6nlXxvzOx3b3OqntLaZEV2svRid0iYILpoxglmjk7k2N4dnP3cmACW1TWw+UsXsMcnExAy+f6W/uWJ2NpfOsqKU9miU0qDTm6ikTBFJsbdjsRzOu7rZ/TXgYhFJtZ3OFwOvGWOKgBoRWWhHI30SeDHSOSrRx/4Sy231ufMmcqisnh1F1SQMgVBVsNYxNLX68Yfka/rkI2v56tObO2VdfdMuxDOhi6d9r8tBZqKHUcleZmQn4XE6+H9Xz2He2FRiBAoqGthZVM2ckMij4UwgSinB69QopSFAbzSGbOAtEdkCrMNyJi8XkS+LSODJf4uIPAwgIrmBbdvpfK/dbx1wT8ARDXweeBjYB+xHHc8nFXuP1eKIET53wSRGJHkwhqGjMbjtDKutltbQ6vNT32yts9heWB08zm+vZJ42MpHLTu3aMXzTgrF8cdHkdoEajhghLd7D23tKaPUb5ow5OQQDQEaCh3uumsmW/Cr+/I6alAaT3kQlbQHmhWm/H7g/TPt64I6Q938B/tLFcd31VSjDiGPVjbyyrYgJGfHEuZ3cvGAcv1qxZ+j4GOzU2w3NPt7aVcK3nt1MVpKXivoWtuRXsnBiOgDHahopqGzg3qWzcB3Hkfo/i6eGbc9IcAcL+8w9iQQDwOWnZvOfWUX89vW9XDRDo5QGC3X/K/2KFdXcPe7+93YKKxv54UetKJ4bFowl3u1gVPLg5kgKEBQMLT4efi+PumYfRXZyu835VcHj8iustrFpcRGNk2mnGB+Z5O3SRzFcUZPS0EAFg9Jv/PX9A0z57ivc8OBq8is6r+oNpaaxhTd2FnNtbk4w5j8jwcO7317EzQvHDcR0T4jXNiVtPlLFxsNWVblA4Z5Nh9uqzAWutWOkUXfJtKvUzRmTHPFcoxk1KQ0+KhiijMLKhuCNp6axhdv/to5tBVWszivj8vvfZd3B8hOcYeBYk1dOq9+wraCKy+9/jzfsnEAAGw9X0BLyNLhixzGaWv3BugsB0uLdQyauPc7WGP76fudaAgWVDcFcR/nllsYQGnnUEwIaw8nkX+jIFbNHcdmpI/nt63uDyQiVgWNo/OKUbvPFJz/kwl+u5EfLd/DIewd4Y1cxf34nj/97aTvbC6u56aE1vLR5aET4Hi6vZ9G0LP79pXMYnRLL7Y+u54G39rH2QDkf+8MH/Pb1vQDUN7fy2zf2MikzntPGpp7grINHwPm8/lAFF57Stn5m8fQRAKzKKwMsU1JmoifimsoZtsYw9ySJSOqKe66yTErffFZNSgPN0PDqKd0mr6SOtHg3j7x/gID5/t+2ILh36Sz+vamQL/9jIyU1Tdx+zoRBm6cxhiPl9ZwxPpXxGfE8/4Wz+PZzW/jFa7uDxzz0bh6v7zzGvuJaWv2GJz+zYEjH7Ife6D9z7kTe31dGs8/PGeNT2XSkguVbivjItCyOVNRHbEYCOHNSOudMzmDu2JNbMARMSnc+uZE/v5PHFy+cPNhTOmlQjSGEnUXVvLWrGGNMj5ym4cgrqaWstmfF3k9EbVMrVQ0t3HbWBP595zlcMTubu6+cCcBtZ43n5gVjeeyO+Zw1KZ3fv7n3uOfaml/Fp/+2LqJSld2hqqGFmqZWxtgOWK/Lwa+umcNNC6wUJjcvHMvUEYmMSoll2XkTefz2BZw1afCqs3WHgPN5fHocZ05KJz3BymiaGufm6tPH8M6eEhb85A3WH6wIpryIhFmjk3n8jgUDWpN5qKImpcFBv3k26w6Wc+2fV2EMzB+fxq6j1az6zkeIjzCG/pZH1nLWpHR+cc2cPptjoN7v6NRYZo1O5vc3ngbAJ07PCcb6e5wOcselsiqvDJ/f4Ah5AjfG8PLWo3xkehYvbS7gzV3FvLu3hItnjuyzOQYIZAoNjcxxOmL48cdO5ZuXnEJyrGtIZEztCSl2zYMb5o9FREhPcFNU1UhSrIu7Lp3G0nmjeGFjIf/dfpRzJw9tIRdNWLmU3uGbz2oupYFi2HzCP315J99/YVvE/R9bdYhEj5OpIxJYe7Cc6sbWiJfmVzW0UFDZEDa/fm8oqAg4NduHMHZcAJYS58YYqG5ovxp35e4Svvjkh7y0qZBNR6womjd2Fkc8n6NVjVz623d5Yk3nDOtBwZDe+ck5Jc4ddUIBrBoK//zcmUETXVq85QsICIxpI5O469JpvPmNC7j2jDFdnkfpGRkJHu61cylplNLAMGwEwzt7S3lq3WEq65tPfHAHNhyq4NVtR/n4aTl859Lpwfa9x2qP06tr9hVb/QoqG7jj0XW8t7c0ovN0pCCgMaQc30yRGm/dqCo6fBbPfmjl9NlaUMUWO+7+9Z3HyCvp3nW+sEu+SoEAACAASURBVLGAWx5Zw6GyOowx3PH3dewsqubVbUc7HfvhIUvwjOmFSWUocsb4tOATa4ZdZzklyqqnRSOXz87mslNH8pvX96hJaQAYNoKhrLaJFp/hlTA3qVA6+g/e3VvCzQ+vITvFy+cvmMSF07LY/5PL8DhjItYY9hVb/fIrGnh9ZzE3P9LdbOSdya+o54k1h3h1WxEFlQ24HFa94OMRqOZVUd+mMVQ1tAQrhb2wqYCmVj+3nTWeFp+fJb99l8dXd1lXCbDKTv7vv7by7t5SPvq79/jDyv1sK7DSQHR0x+wsqubRVQf5xGk5EZviooGAjyE5VgXDQHDPVbNI9Lo0SmkAGBa/Wr/fUFZnPR2/tKmQG+Z3XaPhl//dzd9XHeKBG0+jscXHnU9uZGJmPH+/fT5ZiZaJxhEjTM5KYE9x7zSGUBpbfD0KX1x3sJyH381jxY5jhORsY3RK7AkjdwKF4UO1p/9sKaK51c+s0UlsK6hGBD53/iS+cMEkvvrMJu759w4MMCkzPqwT+P439uI3hqeWLeTuf+/gF6/tJtbl4LRxKZTUtHeyP7shH0eM8P0rpnc6z3BifEY8CR5n8PNW+peASemLT36oUUr9zLDQGCobWvD5DRkJHlYfKOOYXVJxZ1F1pyeLV7YdpaaxlXuX7+D/XtrO5KwEnl52ZlAoBJg6IjGiIuUHS+t4f19Zp/YfvridDYfKySup5ZWtRcc9x7aCKq798yrWHCjns+dP4q1vXMBXPjKFnNRYLumGozjNvlGV17UJhuc+zGdKVgJL51p1j2ZkJzEy2UtWkpe7r5xFs8/P91/Yxm1/WdfpfCU1Tby4qZCrT89h4cR0nv/8Wdx65ji+dtFUxqfHc6ymkZ++vJNn1h+hudXP23tKWDAhLerqEPeU63LH8OY3zo94vYLScy6fnc3lp2arSamfGRYaQ6kdFnrLwnHc9/oelm8pwuuK4bv/2sZvrpvL0nnWzbC51R9clXqgtI5Wv+GmheNIDmMjnpyVwL82FlDd2EJSN9I+Hyyt497lO3hzdzEOET5+2mie/7CARI+TC6Zl8fzGfJ5e31a07q1vXNBlSua1B8oxBl75yrlkJ1vx8F+9aCpfvSh80rWOpNg+hkrblHSgtI4Nhyq469JpwdW0y86b2O5al8wcyavbj9Ls89PQ7Asu5qpramXZY+vxG8OnzracrrFuB3dfZeU5vG/FHirrW4JOwV/9dzfHqpu4/iRwvjodMZ0eKJT+556rZrIqr8yq+PaFs46bqFCJjGHxiZbapoz5E9KYkZ3E0+sO891/WRFKhXaSM4C9xTU0+/ycMzmDVts+Mykz/M05kNUxnFkoHN97YRtrD5TzpQsn8/5di/j2kmkATMtO5Hc3zGPD9y/iN9fN5dJZ1hP/ugPlvLK1KOwq5e2F1WQkeIJCoackepw4YyTofH7+w3xiBD42bzRnjE9j9Xc+wlVz21dM/f2N83jok7kAbMlvy/vz8LsH2Hi4kt/fOI9JmZ0LzmeG+Du+e9l0JmTE43bGcNGMERHNXVFORLptUtpaUMWDGqXULwwPjcE2mWQkuLly7ih+9kpbvaAq+6nZGMPqPCuP0BWzs3lvnxUpNCGj880OYOoIq33vsRocItQ2tXJ2F7Hpu45W896+Ur615BS+cIFl9/T5Dc4YYaJ9/iSvi6XzRnPV3FGcdu8KvvXclmD/2aOTGR+iPWwvrGLW6KSefxA2IkJKnIsH38ljclYCL24q5JwpmcFMnSPDZCt1OmI4fZyVjuK6B1eTFu9mfHoce4/Vsnj6CJbMCl9XIFQw3LRwLJ85byLNrX7czmHxzKEMUS6fnc3LWy2T0uLpIzhlpKbn7ksi+vWKiFdE1orIZhHZLiJ32+0TRGSNiOwVkadFpJORWURuEpFNIS+/iMy1960Ukd0h+7K6M5+AxpCR4AkmYVs0LYtRyV72HKvhy//YyBk/fp17l+8gNc7FhdOy7LnAuDBx9mCFWXpdMew5VssXnviQmx5e0yl5WmOLj1e3HeXlLUXECNwY4vR2xAg/+8Rs7ji3fVoKEWG0nS7h7MnpuB0x/GrFnnbn3Ftcy8xRkQsGsBa6tfoNX3tmM4fL6zlvyokXXKXFu1k0LYuFE9O4ZOYI3M4YRiR7+dpxTFiBvD5AcKWuCgVlILjnqpl4XQ4e0jrRfU6kGkMTsMgYUysiLuA9EXkF+BpwnzHmKRH5E3A78MfQjsaYJ4AnAETkVOBFY8ymkENusov1dJvS2iacMUJyrIvUeDd/vuV0Zo1O5vOPb+Ct3SWAVQDkvKkZnD81i6xEDwkeJ8mxri4dhzGByKRjNdTZVbr+vbkwaGcHuPPJD3l9ZzEjk7yMS4/v5Gy9+vQcwnHnhVN4Ys0h/nTz6fxh5T4eeGs/nz1vIrNGJ7PmQDk+v+l1ScfAmocAM7opaP5y2xk9GudEobOK0l+kJ3iYNSo5WA5W6TsierQzFoH/hst+GWAR8Kzd/iiw9ASnugH4RyRzCKW0ton0BHcwjPOSmSMZnRJLur0ASQR+fd0crjtjLCOTvYgI07MTmZ59/Jvl1KxEdhZVB524+4prg2sgdhRW87q9avhodWNY+3tXLJk1ksduX0C8x8lnz59ESpyL/2cnl3tm3RFS41ycH5K9MxKmdVCtZ2b3T27/gMYw8iQrKKMMDcZnxHOwtG6wpzHsiFjnFxGHiGwCioEVWPWZK40xrfYh+cDorvrbXEdnwfBX24z0felm3oSqhhZSYjuHRgZuWqOSY/E422sGf7z5dH55zezjnnfKiERKay3/xcxRSVQ3tlJiR0D9c8MR3I4Y3HZExKSs8E7sE5HkdfHFCybzzp4SXthYwIodx1g6b3Sn+faUp5edyYqvngfAmLTYsJFXfUGs28Evrp7NPz93Zr+cX1GOx8SMeCrqWyLKeKB0TcSCwRjjM8bMBXKA+UC41UxdpigVkQVAvTEmNMHRTcaYU4Fz7dctXfRdJiLrRWR9SUkJjS1+vK7Ol5JhmznClVjMSPCcMM4+4IAGglXF9hXX0urz8+/NhSyalsWZk6w6v5N7oDF05JYzxzE6JZb/eXoTfmO47azxEZ8rQHKciykjEjllRCKn93ONg2tyxwSzqCrKQBII2jigWkOf0msvoTGmElgJLARSRCTgt8gBjlcx5no6aAvGmAL7bw3wJJbACTfmg8aYXGNMbmZmZperigOmpEhr74YWIg84b/eX1PHuvlJKa5v52Gmjg9FDk7MiFwxel4Pf3zgPl0O4fv4YxqVHpn2E4/E7FnDP0ll9dj5FGUpMyLB+2wfLVDD0JRE5n0UkE2gxxlSKSCywGPg58BZwNfAUcCvwYhf9Y4BrgPNC2pxAijGm1HZoXwG83p35NLb6SQmTryYQShkuw2d3GJ0SS6zLQUOLj3ljU4l3O9hztIb1B8tJjnVxwSmZ5KTGsrOo5oT+ihMxb2wq735rUbvwz76gr8+nKEOJMWlxxAgcKO3bTMYnO5FGJWUDj4qIA0vreMYYs1xEdgBPiciPgI3AIwAiciWQa4z5gd3/PCDfGBMaZ+YBXrOFggNLKDzUnck0NvvwJnW+AabbaZG7Ckk9ETExwpQRCRRUNBDrdjBvbCrv7SulqKqBT5yWg8fpYOao5B5H8nRFuPUFiqJ0jcfpYFRKrDqg+5iIBIMxZgswL0x7HmHMP8aYl4CXQt6vxDI9hR5TB5weyXwaW8Obks6YkMqXFk3mwlO6tRwiLFfMzuZQmfU0snBiGr/8r7Xm4GPzTuRXVxRlIJiQEa+mpD5mWKx8bmzx4Q0TxeNxOvj6xaf06tzLzpsU3F440XI0j0mLDa4SVhRlcBmfHs8LmwowxkRlAaihyLBYotpVVFJfMzsnhcxEDzfOH6dfQEUZIozPiKemsbVdNmGldwwLjaGhxYfX3f+pj93OGN779oXBtQuKogw+oZFJ6QkabNEXDIs7XHOrP6wpqT/wOB2qLSjKECKQCFMjk/qOqBcMgepmWixFUU5ORqVY0XxHqxpOcKTSXaJeMARyFw2Ej0FRlKGHx+nA7YyhprH1xAcr3SLq76b+oGBQjUFRTlaSvE6qbcHQ2OLjv9uPBh8alZ4zDASD9TdWBYOinLQkel3UNFpZkJ9ed4Rlj21g45HKE/RSuiLqBYOakhRFSfQ6g6akDYcqAFi5y0qLv/FwBW/tLh60uUUjUX83DWgMHtUYFOWkxRIMlsaw8YglGN60hcF9r+/lS09upK5JfRDdZRgIBltjGKBwVUVRhh6JHhc1ja2U1DRxpLyBjAQ32wqqqWlsobi6kdqmVpZvOV6yZyWUqBcMJhiuGvWXoihKhARMSRsPW9rCRTNGAlBW20yJXRP+ybVH2vXZX1KrDuouiPq7aZldUS12AFY+K4oyNAk4nz88XInLIZw/1aqfUlzTRHl9M5mJHjYfqWRHYTUAh8vqWfzrt3lzV5vvwRhDq88/KPMfakS9YKix7YZqSlKUk5dEr5O6Zh8bDpUzIzuJ7ORYAPYcq8EYuO2s8bidMTy17jAA+ZX1GGNpDQF+8vJOFv/67X7XIv6wch/Pbcjv1zF6S9QLhgC6jkFRTl4SvVbat42HK5k3NpU0u3rj7qM1gFVh8bJZI/nXhwU0NPuCCfcKKxsBOFhax1/fP8jBsnp2H6vpt3m2+Pz87o19PL9RBcOAoD4GRTl5SfJaFRxb/YZ5Y1NIT7AEw66jlukoK9HDDfPHUtPUyr+3FIYIBiuNxi//uxtHjJUD7b29pf02zy35VTS0+CirHdqZYCO+m4qIV0TWishmEdkuInfb7RNEZI2I7BWRp0XEHabveBFpEJFN9utPIftOF5GtIrJPRO6XbmasU41BUU5ekmLbEkWfNjaVOLcTryuGXbbGkJnoYf6ENE4Zkcif394fdEgXVTWyNb+K5VuKWHbeRCZmxvPevv4TDGsOlAFQGkYw7DlWw+ce20Bji6/fxu8uvXnMbgIWGWPmAHOBJSKyEKv2833GmClABXB7F/33G2Pm2q/PhbT/EVgGTLFfS7ozGY9TNQZFOVlJtDWGjAQPOamWfyE93hNc9JaZ6EFE+J/FU9hfUsdjqw8Blsbw81d3kRrnYtl5E5mbk8Keo/1nSlqdVw5AeV0Tfn97X8Yz647w6vaj5JUMfjW6iO+mxiLguXHZLwMsAp612x8Flnb3nCKSDSQZY1YZywP09+7211TYinLyEvAxnDY2JXgvCPgZkmNdeOzglEtmjiQ51kVlvbUYrqyumff2lXLnoikkel2kxrupbGjplzm2+PxsOFiOxxmD39BpnICmUlzT2C/j94RePWaLiENENgHFwApgP1BpjAksMcwHuiqOPEFENorI2yJyrt022u4T4Hj9FUVRgDYfw7yxbSV3A4JhzpiUYFtMjDA5K6Fd39Epsdy8cCwAqXEu6pt9NLX2vTlnW0EVdc2+YA368rqm4L6Smqag2au4pils/4GkV4LBGOMzxswFcoD5wPRwh4VpKwLGGmPmAV8DnhSRJCDcY3+n/iKyTETWi8j6TFcLTy1bGPlFKIoS9YxLj+MHV8zghvljgm2BG+ySmSPbHTspMx6AEUlWtbdvXnJKUKNIjrOESVV9C3uP1URcLnTPsRou+vXbHClvKx605oBlRrr0VGs+oX6GD/a3+TVKol0wBDDGVAIrgYVAiogEPEE5QKd16MaYJmNMmb29AUvTmIqlIeSEHNpV/weNMbnGmNyRWRksnJjeF5ehKEqUIiJ8+pwJpMS1xbok2w7pi2aMaHfspExLY8gdl8brXzufpfPajBKpcZbmUdnQwh1/X8/tj66LaF3Db9/Yy97iWlbtLwu2rc4rY3JWAtNGJgG0i0z6YF8ZSV4nCR5ndAsGEckUkRR7OxZYDOwE3gKutg+7FXixi74Oe3silpM5zxhTBNSIyEI7GumT4foriqKciPuvn8fjty8gM7F9HeiJtmCoqG/uZFZKibUES0VdM5X1LWw8XMlLm3uWY+lAaR2vbC0CYEeRFS7b6vOz/mAFCyakBUNpy+qaqGpowe83vLevlLMmZZCV5KG4ppFXtxXxm9f39Pyi+wjniQ/pkmzgUfsGHwM8Y4xZLiI7gKdE5EfARuARABG5Esg1xvwAOA+4R0RaAR/wOWNMuX3ezwN/A2KBV+yXoihKj8hK8pKV5O3UHohaSo51ddqXEqIxBMJGf/7KLi6ZOfKEIfHf+Odmzp2Sweq8MpyOGEanxAbXUWwvrKa2qZWFE9NJjXMjAtsLqvnBi9u5acFYCiob+NwFk6hsaOZoVSOfe/xDAO68cDJOR9vze3VjC4keZ78H20QsGIwxW4B5YdrzsPwNHdtfAl6yt58DnuvivOuBWZHOS1EU5XhMG5nI/7t6Nounj+i0LyAYKuqaaWr1s2BCGmsOlPPwu3ncuWhKl+dsbvXz7IZ8nt2Qj9sRwzW5OfgNvLy1iHf3lnDLI2sBWDAxDUeMkB7v4en1VlK/J9ZYaTrOmZzBugPl7TSUQ+X1QdPX0apGFv70DX5wxQw+fc6EvvkwukCD/xVFOakQEa7NHROMWgol1fZRBBzX55+SycUzRvCHlfspru46jPRYyL5Wv5/PnjeJGdmJVDW08NnHNpCZ6OGa03PISrQ0mC9cMKld/1HJXsanx3WqRBm6pmJHURUAr+881pPLjQgVDIqiKDZxbgcuh3DUvtF7nA6+dvFU6pt9x60CV2Cn1gC4YvYoxqbH8dE5ozhvaiZxbifPfPZMfnHNnOAxnzp7PL+9fi5fv2gqAGdPzkBEOH28FW77j88sRIRg3qZj1Y3sPWYtGwsn0Pqa3vgYFEVRhhUiQkqcm6NVlmDwumIYn26Ft4ZLYwFQ09jC3mLrpn3vVTO5YvYoAFLi3Pz9052s6sFxrpo7mqNVjTywch+X2CG11+aO4co5o/C6HIxLi+O5D/N5aXNhu9XQA1FBQgWDoihKCKlxrjbB4HTgdTlI9DgprQ0fRnr739az9qAVO3NN7pge5W0bmexl8w8vDq6jgLa8b2eMT+PFzYWcOTGdJK+LTUcqASgfgAR8KhgURVFCSIl1s8+u0xC4SacnuMNmRC2ubgwKBbcjJqJknp4uasn8/BOz+dHHZuFxOto5sMvq+n+dg/oYFEVRQkj0OoMrngPp/NPi3WFvyKF+h+Y+rv4WEyNBoXHWpAyuPj2HU0YkDkjKbhUMiqIoIcR72gwpgRtzeoIneEMOTYu9cncJ2clWpJHb0X+3U0eM8Mtr5nDJrJGU1zf3ewlSFQyKoighhAqGgMaQkeCmtLaZJ9ccZtr3Xw36IDYermTBhDSe+eyZvPyVc8Oery/JSHBjDFTU908G2ADqY1AURQkhwdPZEZwe76G8rom/fXAAgMKqBmIEjlY3MjsnhfkT0gZkbhkJVnqPsrqmTqk++hLVGBRFUUKIc3fWGNIT3PiNlQcJoLaxlc351oKzOWOSB2xu6fYahv72M6jGoCiKEkJCFz4GgBaftYqgor6ZvcdqccQIM7IHTjAEFrdFmg68u6jGoCiKEkJ7H4MlGDI6rDaurG9hc34lU0ckEuseuHrzgRKmgZKl/YUKBkVRlBDiQ3wMHtuUND07qd0x5XXNbC2oYk7OwGkL0FbCtKaxf53PKhgURVFCCDUleW1TUmq8m6tPb6shtrWgisr6FmbnpHTq35/EuR04YkQ1BkVRlIEkYEqKEXA52uoe/ORjp/Knm09nTFpssBTn7AHWGESEBI9TNQZFUZSBJKAxeF2OdgVx3M4YlswaSVqcm8YWPx5nDKeMTBzw+SV6nUNTYxARr4isFZHNIrJdRO622yeIyBoR2SsiT4tIp/ywInKRiGwQka3230Uh+1aKyG4R2WS/siK/NEVRlJ4THyIYwpFqO6JnjErC1Y+rnbsi0euieigKBqAJWGSMmQPMBZaIyELg58B9xpgpQAVwe5i+pcBHjTGnYtWEfqzD/puMMXPtV9cJ0BVFUfqBeDvKyOMMf3sMFPOZM8D+hQCWxjAETUnGotZ+67JfBlgEPGu3PwosDdN3ozEmULtuO+AVkf5bwqcoitIDTqQxBMp/DrR/IUDSUDUlAYiIQ0Q2AcXACmA/UGmMCcw4Hxh9gtN8AthojAlNW/hX24z0fenviteKoigdiHM7EDmxxjDQEUkBEr0uapqGaK4kY4wPmCsiKcC/gOnhDuuqv4jMxDI9XRzSfJMxpkBEEoHngFuAv4fpuwxYBjB27NhIL0FRFKUTIkK829mlxnD57Gxa/YaJGfEDPDOLIet8DsUYUwmsBBYCKSISEDY5QGG4PiKSgyVMPmmM2R9yrgL7bw3wJBC2Lp4x5kFjTK4xJjczM7O3l6AoitKOeI8jmCepI5MyE/jaRVOJiRkcg0ZAMBhjMMZQ29T3QiLSqKRMW1NARGKBxcBO4C3gavuwW4EXw/RNAf4DfMcY835Iu1NEMuxtF3AFsC2S+SmKovSGeE/XGsNgk+h14fMb6pt9fP2ZzZz78zfx+/u2EnSkpqRs4FERcWAJl2eMMctFZAfwlIj8CNgIPAIgIlcCucaYHwB3ApOB74vI9+3zXQzUAa/ZQsEBvA48FOH8FEVRIubKOaOCKa6HGoG0GD99ZSfPbywAoLqxhZS4TqsDIkaM6VtJM9Dk5uaa9evXD/Y0FEVRBoSXNhfy5X9sBGBUspfCqkbe+sYFTOihz0NENhhjcsPt05XPiqIoUURAY8gdl8oPr5wJ9H0abq3HoCiKEkWcMT6Nz18wiTvOmcCRigYAKutVMCiKopy0JHicfHvJNIBgRFJf14BWU5KiKEqUEsjbVGGbklp8/j6JUFLBoCiKEqUkepw4Y4QK25T0qb+u46oH3udYdSMATa2+iM6rgkFRFCVKERFS4lxBU9Le4hq2FlSx9IH3eWzVQU753qu8u7ekx+dVwaAoihLFpMa5qaxvxhhDeV0zi6dnYQx8/8XtAGwrqO7xOVUwKIqiRDGpcW7K65qpbWqlxWdYMCGdf33xLC6ZOQKAVp+/x+dUwaAoihLFpMS5qKxvCa5lSI13k50cy59vySXO7aCyoecRSyoYFEVRopi0eDfl9c2U2YIhPb4tNUZyrIuqCASDrmNQFEWJYrKSvJTVNlFcbZW1SetCMBhj2Fdcy+q8smCYa1eoYFAURYliRqd48RvYUWQ5mUMFQ5ItGLbmV/HpR9dRUtPU1WnaoaYkRVGUKGZUSiwA2wuqAEhPaK8xVDe0sOlIBSU1TfzfR2ew8hsXcP0ZY457ThUMiqIoUUxAMGwtqMLjjCE2pI5EwJRUY6fOuH7+WMZnxPOTj5163HOqYFAURYliRiVbgqG4pon0eDcibZXlAoKhrqkVR4wE61ifqPqcCgZFUZQoJtbtCPoV0hLaO5WTY13UN/uoqG8hweNsJzSOR8SCQUS8IrJWRDaLyHYRudtunyAia0Rkr4g8LSJh3d8i8h0R2Sciu0XkkpD2JXbbPhG5K9L5KYqinCw0NFs5keaPT2/XnhzrAqCwsoEET/djjXqjMTQBi4wxc4C5wBIRWQj8HLjPGDMFqABu79hRRGYA1wMzgSXAH0TEYZcKfQC4FJgB3GAfqyiKonTBqBQvAF+8cFK79oBgKKjomWCIOFzVWDVBa+23LvtlgEXAjXb7o8D/AX/s0P0q4CljTBNwQET2AfPtffuMMXkAIvKUfeyOSOepKIoy3Pnbp+ZT1dBCeoc61aEawykjE7t9vl75GOyn/E1AMbAC2A9UGmNa7UPygdFhuo4GjoS8DxzXVXvHcZeJyHoRWV9S0vPMgYqiKMOJMWlxzBqd3Kk9Oc4SDHXNPhK8rm6fr1eCwRjjM8bMBXKwnvinhzssTFs4D4g5TnvHcR80xuQaY3IzMzN7MmVFUZSThswQDSLB4zjOke3pk5XPxphKEVkJLARSRMRpaw05QGGYLvlA6AqL0OO6alcURVF6QEY7wTAAzmcRyRSRFHs7FlgM7ATeAq62D7sVeDFM95eA60XEIyITgCnAWmAdMMWObHJjOahfinSOiqIoJzOxbgeJtkCIHwjnM5ANPGpHEsUAzxhjlovIDuApEfkRsBF4BEBErgRyjTE/MMZsF5FnsJzKrcAXjTE++7g7gdcAB/AXY8z2XsxRURTlpCY9wU1NU2tQQHSH3kQlbQHmhWnPoy3CKLT9JUKe/o0xPwZ+HOa4l4GXI52XoiiK0kacu+cag658VhRFGcbEuS2nc4JXBYOiKIqC5WeAAXI+K4qiKEMfr51t1eXo/u1eBYOiKMowJpCGO5BPqTuoYFAURRnGXDlnFEDYldFdoaU9FUVRhjGLZ4xg748vVVOSoiiK0kZPhAKoYFAURVE6oIJBURRFaYcKBkVRFKUdKhgURVGUdqhgUBRFUdqhgkFRFEVph1ilm6MXEakBdg/wsBlA6TAaR8fT8aJhTL2+vmWcMSZsCczhsMBttzEmdyAHFJH1AzHmQI2j4+l40TCmXt/AoaYkRVEUpR0qGBRFUZR2DAfB8OAwHnOgr03H0/GG8ph6fQNE1DufFUVRlL5lOGgMiqIoSh+igiEMIiLDaZzBHlOJXgb6+zLcv5/Rcn0qGMIzUP+8YLjwAH5hUuzx+j1UWUROEZEB/Y6JyCIRGTlAY90oInPs7YF6mEgJ2R6IMQf6HuENbETLTbSHuAd7At1hSAsGEVkqIvcO4HiXiciLwC9E5IJ+HGeJiLwG/FJEPgZg+tnZIyLJIvJf4FV7vNZ+HOsiEVkD3MEAfcdE5CwR2Q7cBiT081iLReRd4DfAPBiQ/9+lIvI28ICIfKe/xxSRy0VkOXCviJzdX+OEjHexiHwA/F5EboJ+v76lIvI7EUnrrzE6jHeZiLwK/FZEbhmIMXuFMWZIvbCe1h1YN5V9QAtwbj+P5wZ+BawFLgV+APwZWNAP4/wSeB+4EvgC8CQwZQA+Vw/wNFAHXGO3Ofr4+lzAPcBe4OMd9/fjnwT/8QAADK5JREFUtTmAvwI39vP3JBZ4BlgJXAL8AfhCX3+WYcaeD6yxvzPnA/8EZvXjeKfbv4XLgOuxomVus/fF9MN4mcBqYCnwEeAl4Dv9MZ79f/w4sBPIBz7RH9cUMp4T+F9gHbAEuAn4B/DR/hqzL15DTmMwFj4soTAP6+bZb1qDPV4zsAfrxvIK8DCWyaX71bO7P86rwPnGmJeAD7AE34G+GiccIuLAup7VwHXA7+w5+fpKXbevrwXwA88aY563xz5XRFx9McZxSML6wb8sIm4RuUVEJouI255Dr6/Rvr4G4AljzAXGmNew/n+32Pv77LsShrOBd+zvzBGs7+X+gJmuH0wui4F3jTEvAy8CR4EviUiyMcbfl+PZ5xoBbDbGvGCMeQO4C/iGiGT09XjGulvnAecAXwFuBnL66vxhxmu1x7veGPMqltArZIiblIaMYBCRL4vIQyJyh930tjGmxhjzEBAvIrfbx/XJnEPG+4zd9BCQJyJuY0whkAik9+E4dwAYY143xrSKyGXA88BU4Ccicp19fK9/BCFjflpExL5pVQOXG2OWA1tE5AciMssYY3ozZshYy+ymPwHZIvJXEdkKfAt4BPi0fXxfXt/tdlMMMBGYjfU0/VHgJ1haH/TCZ9Txe2KMedFud2AJ9O0iMibS83dnTOB14EYR+R3wDjAK+CNwdz+N9xZwhYik2sKwBev78y3ovYlHRG4VkYtCzlULnBUw6xhjdmD9H3/Xm3HCjWezzRhTZox5DuvaPh54iOin8Z4HDoiIyxhTgyWI4vpqvH5hsFUW+zt2G9bT7BLgbeA7wKSQ/ZcC24HUfhxvcsj+VOANYGQfj/O/gXGwzANT7e3LgNeA8f1wbf8LTAKygB/Zx3waaAXW2+9dfTTW9+zPbinwBDAN66Z8FfAfYGw/XN/3sEw8PwP2A9fZxyUAJUBuH3+WE0P2n4plIkjsx9/C97G0vVTg19gmCGA6sA2Y2cfjfdf+rvwOWA68i2WmuwRLGMX3YqxU4FmgCNhCiPkN+DvwWIdj1wAT+no8rAeJwBqus7F+66d16Ntj0+fxxgs5xgu8AJzSV9+Z/ngNFY3hI8DPjaVqfR3rw7spsNNY5p2dwDIRSRSRa/phvBtD9o8HqowxR0UkR0QW9dE4buzrMsasNcbssY/bgXUT6wuHcLhruwZoAC61HdBfBt4EDtl9Ih2341ge4LPGmBeAZcaYXcb6NWwBKrGeznpLuOv7ApZfKN5+YYypBZ7C+rH21VhuLNMD9hhbsT7X63sxxonGdAF3GmMqsLTLwP9sF7AK6zPvy/G8wCeNMV/C+lzvMcZ8CmgEvMaYukgHsq/hv1hCbQPW/yzAncASETnDfl8HbAaa+2M8+3uJMeZ9YBPWb2NaQPMN7O+r8UJIwfocd4vIGBH5RE/HGQgGVTCEmIU2AlcAGGPWY33hR3WIhvg28FMsx2ZE4YjdGO9ce/9owCEiX8J60u3ReMcZZzWWmaVjlMdtWKplWU/G6eaYHwATsGyqK4C1xpj/3975h9pZ13H89d6P1Lk5jdpwmpuy6dTQwSQU5zYCh21KYZqkqJkUDQpKFApmQSz68YeU3DQtYlpkSEsJQlRo/ti1kWibbohKGEsXbpa6zXmN3fvpj8/32T3nsu49z/nxved7+7zg4d5zzvM87/P+nnOez/P9fH8tM7M1wGpJp9f9EYyjNQicLuniMReQG/G7+rfbMje+5lbgHOBkPNVxmaQrJG3A7wZf6qLWNhq+lykt9hhwbKcpsgk+v0WSzsGD+S8kzcJrSh/HG1C7qTcILJG0wsx2m9njab91eI2sLRrK534zewdvuL9S0sKkvR9Pjd0u6UZG/R3stp55u8X0hjL4MZ41eBKvLdVOebagV3UPPwOYI+nreHvDUae9nmxy9zGfnv4KwMxG0kuDwDRJK9PjnXh1bEHafzFe0A/jVb6Wco9t6FUB4FI8T70YWGtmv+mRrxsk7cQv3OvN87ktUUNzF/AG3mbybTPb0HCa08xswobvmv72NPj7rKQd+I9hvZkN9cDfTvziuNzM7sfbOFYApwGXm9mEF852P78UUOcB77URXOv6W2pmd+Brj/wOD4ZXmtneHujtwQMtklbKu8kuwcu2XX/VHfpQ+vss8AjwveoYMxvAL9LLgYXAVWb2bi/0zGw4XbDnAwN40F1mZhsbj++iXlUrXw5chF9b1plZy2WakyzrMUi6CM9rvyzp3nR3gKQZqcBexS9g10gaNLPX5YOUDqVTvItXp1/tsd4H6RSbgT+Y95DohU4VAF7AUy7PtOKrTc1/SFoAvG9m/0lfaDOzkYnSAl3w9wrwFTP7cw/9vS5pHn7hwsz+JOmJhgtfL/wdajjNrea9zXrpbz5wVjrFzcCsdFfaK73Gz/DveJfcXR3qHfneNew+ADwo6Vx8gZo56fN70lrs5dWB3l68HWo38DUz253B32G8YX+VmT3dit5k0fMaQ7oTqSLyAuBbktZAUxQ9gDdyfQgf9DUTzw3/K+23r0ZQ6ERvb9rvqRaCQic6b6X9ttcMCu1qnshoWQ63eNHshr8XawaFTjT3VefJ4O9Iyq9mUOhE881Kr0ZQ6MZnuLtGUBhPr7pDP07S7OrcwEPAi3ga54Rq3wx6T+OdWYZrBIVO9J7CV0zb2e9BAeh9ryTgFuDO9P+HgV/ivRzmp+c24l3TluLV1014Lvce2hg0lEsvt6/cmuGvbH99qvddvOvmeenx5/HG9B/RRs+4qa43mVvXU0mSLgT+baM9bl4Gzpe0wMz2SDqIr236aUlP4Pnnb5rZ39LxX8S7xB3oJ73cvnJrhr+y/RWitxi4rdLDx4Gsthbauf4f9PqJrqWSJJ0o6Y94z5fPVdUpPIe5H9gkaTPwMbwnxAlm9oqZXWtmR0Zxmue+W/mhZ9HL7Su3Zvgr21+hetOT3rZWLppTXa8f6dpCPZJOwecd2Y83lFVD6pGPKrwYr3L9VtKn8Mbkden1adZCXngy9HL7yq0Z/sr2F3rl6/UjHQUGSTfgObS/mtl+ScfitZDbAAH3mk8vMfa4DcA75t3T+k4vt6/cmuGvbH+hV75ev1M7lSTnZElb8IFL1wF3yye8GjKzQ/jcLicBnxxz7ApJzwGX4MPt+0Yvt6/cmuGvbH+hV75eUVi9Vvlq7o8zgV+n/2fg86r8fsy+38B7Pcwlza+Cd/Fa2296uX3l1gx/ZfsLvfL1SttaLcQZ+GyVP8Tng78CuK/hdeEjQlc1PDcbH8X4F7wP9qk1PrQserl95dYMf2X7C73y9UrdWinIVfhkVncDX8IHalyGjxj8RMN+64EtDY+vwSfA+jkwr8YHl0Uvt6/cmuGvbH+hV75eyVsrhXkJcH3D47tSwX0BeC49Nw2fZ+hB0tTR+FTLK2u/oUx6uX3l1gx/ZfsLvfL1St5aKcxZ+NS+VU7uOuD76f/t+DwjABcAD3T8hjLp5faVWzP8le0v9MrXK3mbsFeSmR0ysw9sdP6SSxmdl+Ym4Gz5ouEPAM+Dt/ZPdN7J1svtK7dm+CvbX+iVr1c0NaLtdLya9Qijq5AtxidoWwGc0s2IlUsvt6/cmuGvbH+hV75eiVudcQwj+GpSbwHnpch6OzBiZlvN7I0a5+onvdy+cmuGv+4zVX8Lode770xZ1Iy0F+KFuhW4uddRK5debl+5NcNf+ZqhV7ZeaVvdwjwVXwLvmCxvLpNebl+5NcNf+ZqhV7ZeaVvXJtELgiAIpgY9X8EtCIIgKIsIDEEQBEETERiCIAiCJiIwBEEQBE1EYAiCIAiaiMAQBDWRNCxpu6RdknZIukVp3eRxjlkk6dpc7zEIOiECQxDU530zW2Zm5+Lz7awFvjPBMYuACAxBEcQ4hiCoiaSDZja74fEZwLPAR4CFwK+A49PLXzWzZyRtA84GXgPuA+4EfgCsxmf8/KmZ3ZPNRBCMQwSGIKjJ2MCQnnsbWAocwOfcGZK0BJ+++QJJq4FbzezytP+X8UVfNko6BhgErjaz17KaCYKjMGOy30AQTBGq6ZlnAgOSlgHD+JrCR2MNPoHbVenxXGAJXqMIgkklAkMQdEhKJQ0De/G2hjeB8/E2vKH/dRi+MMyjWd5kENQgGp+DoAMkfRT4GTBgnpedC/zTzEaA6/G5/8FTTHMaDn0UWC9pZjrPmZKOJwj6gKgxBEF9jpO0HU8bHcYbm+9Ir90FbJZ0NbAFeC89/wJwWNIOYBPwE7yn0vNplbB9wGdyGQiC8YjG5yAIgqCJSCUFQRAETURgCIIgCJqIwBAEQRA0EYEhCIIgaCICQxAEQdBEBIYgCIKgiQgMQRAEQRMRGIIgCIIm/gvgCRkQ8PM+/wAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 時間條件式搜索\n",
    "twdusd['2019'].twdx.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Date\n",
       "2019-12-26    29.900000\n",
       "2020-01-01    29.898001\n",
       "2020-01-02    29.844000\n",
       "2020-01-03    29.900000\n",
       "2020-01-08    29.985001\n",
       "2020-01-10    29.943001\n",
       "2020-01-13    29.962000\n",
       "2020-01-14    29.874001\n",
       "2020-01-15    29.826000\n",
       "2020-01-16    29.933001\n",
       "2020-01-17    29.899000\n",
       "2020-01-20    29.950001\n",
       "2020-01-21    29.891001\n",
       "2020-01-23    29.919001\n",
       "2020-01-24    29.975000\n",
       "2020-02-06    29.951000\n",
       "2020-02-12    29.948999\n",
       "2020-02-13    29.910000\n",
       "2020-02-14    29.985001\n",
       "2020-03-03    29.805000\n",
       "2020-03-04    29.770000\n",
       "2020-03-05    29.750000\n",
       "2020-03-06    29.884001\n",
       "2020-03-09    29.900999\n",
       "2020-03-10    29.985001\n",
       "2020-03-11    29.740000\n",
       "2020-03-13    29.948000\n",
       "2020-04-13    29.850000\n",
       "2020-04-15    29.905001\n",
       "2020-04-16    29.937000\n",
       "                ...    \n",
       "2021-01-29    27.932600\n",
       "2021-02-01    28.000999\n",
       "2021-02-02    27.950300\n",
       "2021-02-03    27.858000\n",
       "2021-02-04    27.864000\n",
       "2021-02-05    27.902901\n",
       "2021-02-08    27.973600\n",
       "2021-02-09    27.880899\n",
       "2021-02-10    27.853901\n",
       "2021-02-11    27.961300\n",
       "2021-02-12    27.943100\n",
       "2021-02-15    28.014999\n",
       "2021-02-16    27.993999\n",
       "2021-02-17    28.040001\n",
       "2021-02-18    27.918699\n",
       "2021-02-19    27.860800\n",
       "2021-02-22    27.907801\n",
       "2021-02-23    27.809401\n",
       "2021-02-24    27.850201\n",
       "2021-02-25    27.650000\n",
       "2021-02-26    27.764999\n",
       "2021-03-01    27.855000\n",
       "2021-03-02    27.819000\n",
       "2021-03-03    27.740000\n",
       "2021-03-04    27.719999\n",
       "2021-03-05    27.813000\n",
       "2021-03-08    27.927500\n",
       "2021-03-09    28.283001\n",
       "2021-03-10    28.252001\n",
       "2021-03-11    28.304001\n",
       "Name: twdx, Length: 256, dtype: float64"
      ]
     },
     "execution_count": 98,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 一般條件式搜索\n",
    "test = twdusd[twdusd.twdx<30].twdx\n",
    "test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                   twdx   R-squared:                       0.865\n",
      "Model:                            OLS   Adj. R-squared:                  0.865\n",
      "Method:                 Least Squares   F-statistic:                     3532.\n",
      "Date:                Thu, 11 Mar 2021   Prob (F-statistic):          8.21e-242\n",
      "Time:                        09:12:20   Log-Likelihood:                -260.36\n",
      "No. Observations:                 553   AIC:                             524.7\n",
      "Df Residuals:                     551   BIC:                             533.4\n",
      "Df Model:                           1                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "const         31.6011      0.033    958.519      0.000      31.536      31.666\n",
      "t_trend       -0.0061      0.000    -59.434      0.000      -0.006      -0.006\n",
      "==============================================================================\n",
      "Omnibus:                       13.182   Durbin-Watson:                   0.062\n",
      "Prob(Omnibus):                  0.001   Jarque-Bera (JB):                7.800\n",
      "Skew:                          -0.109   Prob(JB):                       0.0202\n",
      "Kurtosis:                       2.460   Cond. No.                         637.\n",
      "==============================================================================\n",
      "\n",
      "Warnings:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
     ]
    }
   ],
   "source": [
    "x = sm.add_constant(twdusd['t_trend'])\n",
    "result = sm.OLS(twdusd.twdx,x).fit()\n",
    "print(result.summary())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x261814fab70>"
      ]
     },
     "execution_count": 100,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "result.resid.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [],
   "source": [
    "twdusd['q_dum_1'] = 0\n",
    "twdusd['q_dum_2'] = 0\n",
    "twdusd['q_dum_3'] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\queti\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:2: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame\n",
      "\n",
      "See the caveats in the documentation: http://pandas.pydata.org/pandas-docs/stable/indexing.html#indexing-view-versus-copy\n",
      "  \n"
     ]
    }
   ],
   "source": [
    "# 季節性dummy\n",
    "twdusd.q_dum_1[twdusd.index.quarter == 1] = 1\n",
    "twdusd['q_dum_2'][twdusd.index.quarter == 2] = 1\n",
    "twdusd['q_dum_3'][twdusd.index.quarter == 3] = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                   twdx   R-squared:                       0.133\n",
      "Model:                            OLS   Adj. R-squared:                  0.128\n",
      "Method:                 Least Squares   F-statistic:                     28.11\n",
      "Date:                Thu, 11 Mar 2021   Prob (F-statistic):           6.42e-17\n",
      "Time:                        09:29:58   Log-Likelihood:                -774.67\n",
      "No. Observations:                 553   AIC:                             1557.\n",
      "Df Residuals:                     549   BIC:                             1575.\n",
      "Df Model:                           3                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "==============================================================================\n",
      "                 coef    std err          t      P>|t|      [0.025      0.975]\n",
      "------------------------------------------------------------------------------\n",
      "const         29.3880      0.088    334.683      0.000      29.216      29.560\n",
      "q_dum_1        0.3503      0.114      3.064      0.002       0.126       0.575\n",
      "q_dum_2        1.0568      0.123      8.560      0.000       0.814       1.299\n",
      "q_dum_3        0.7363      0.127      5.818      0.000       0.488       0.985\n",
      "==============================================================================\n",
      "Omnibus:                      385.758   Durbin-Watson:                   0.014\n",
      "Prob(Omnibus):                  0.000   Jarque-Bera (JB):               44.091\n",
      "Skew:                          -0.339   Prob(JB):                     2.66e-10\n",
      "Kurtosis:                       1.794   Cond. No.                         5.03\n",
      "==============================================================================\n",
      "\n",
      "Warnings:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
     ]
    }
   ],
   "source": [
    "x = sm.add_constant(twdusd[['q_dum_1','q_dum_2','q_dum_3']])\n",
    "y = twdusd.twdx\n",
    "result = sm.OLS(y,x).fit()\n",
    "print(result.summary())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Date\n",
       "2018-12-31    1.165011\n",
       "2019-01-01    0.826663\n",
       "2019-01-02    0.762662\n",
       "2019-01-03    1.020664\n",
       "2019-01-04    1.124664\n",
       "2019-01-07    1.044664\n",
       "2019-01-08    1.064662\n",
       "2019-01-09    1.077663\n",
       "2019-01-10    0.972663\n",
       "2019-01-11    1.040662\n",
       "2019-01-14    1.047662\n",
       "2019-01-15    1.095663\n",
       "2019-01-16    1.089662\n",
       "2019-01-17    1.026662\n",
       "2019-01-18    1.088663\n",
       "2019-01-21    1.114664\n",
       "2019-01-22    1.137662\n",
       "2019-01-23    1.138664\n",
       "2019-01-24    1.119663\n",
       "2019-01-25    1.135663\n",
       "2019-01-28    0.969663\n",
       "2019-01-29    1.049663\n",
       "2019-01-30    1.042663\n",
       "2019-01-31    0.984662\n",
       "2019-02-01    0.947664\n",
       "2019-02-04    1.024663\n",
       "2019-02-05    1.038663\n",
       "2019-02-06    0.990663\n",
       "2019-02-07    1.048664\n",
       "2019-02-08    1.069663\n",
       "                ...   \n",
       "2021-01-29   -1.805737\n",
       "2021-02-01   -1.737338\n",
       "2021-02-02   -1.788037\n",
       "2021-02-03   -1.880337\n",
       "2021-02-04   -1.874337\n",
       "2021-02-05   -1.835436\n",
       "2021-02-08   -1.764737\n",
       "2021-02-09   -1.857438\n",
       "2021-02-10   -1.884436\n",
       "2021-02-11   -1.777037\n",
       "2021-02-12   -1.795237\n",
       "2021-02-15   -1.723338\n",
       "2021-02-16   -1.744338\n",
       "2021-02-17   -1.698336\n",
       "2021-02-18   -1.819638\n",
       "2021-02-19   -1.877537\n",
       "2021-02-22   -1.830536\n",
       "2021-02-23   -1.928937\n",
       "2021-02-24   -1.888136\n",
       "2021-02-25   -2.088337\n",
       "2021-02-26   -1.973338\n",
       "2021-03-01   -1.883338\n",
       "2021-03-02   -1.919337\n",
       "2021-03-03   -1.998337\n",
       "2021-03-04   -2.018338\n",
       "2021-03-05   -1.925337\n",
       "2021-03-08   -1.810837\n",
       "2021-03-09   -1.455336\n",
       "2021-03-10   -1.486336\n",
       "2021-03-11   -1.434336\n",
       "Length: 553, dtype: float64"
      ]
     },
     "execution_count": 131,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result.resid"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "twdusd."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x261831ba2e8>"
      ]
     },
     "execution_count": 130,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "adj_y.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Date\n",
       "2018-12-31    1.165011\n",
       "2019-01-01    0.826663\n",
       "2019-01-02    0.762662\n",
       "2019-01-03    1.020664\n",
       "2019-01-04    1.124664\n",
       "2019-01-07    1.044664\n",
       "2019-01-08    1.064662\n",
       "2019-01-09    1.077663\n",
       "2019-01-10    0.972663\n",
       "2019-01-11    1.040662\n",
       "2019-01-14    1.047662\n",
       "2019-01-15    1.095663\n",
       "2019-01-16    1.089662\n",
       "2019-01-17    1.026662\n",
       "2019-01-18    1.088663\n",
       "2019-01-21    1.114664\n",
       "2019-01-22    1.137662\n",
       "2019-01-23    1.138664\n",
       "2019-01-24    1.119663\n",
       "2019-01-25    1.135663\n",
       "2019-01-28    0.969663\n",
       "2019-01-29    1.049663\n",
       "2019-01-30    1.042663\n",
       "2019-01-31    0.984662\n",
       "2019-02-01    0.947664\n",
       "2019-02-04    1.024663\n",
       "2019-02-05    1.038663\n",
       "2019-02-06    0.990663\n",
       "2019-02-07    1.048664\n",
       "2019-02-08    1.069663\n",
       "                ...   \n",
       "2021-01-29   -1.805737\n",
       "2021-02-01   -1.737338\n",
       "2021-02-02   -1.788037\n",
       "2021-02-03   -1.880337\n",
       "2021-02-04   -1.874337\n",
       "2021-02-05   -1.835436\n",
       "2021-02-08   -1.764737\n",
       "2021-02-09   -1.857438\n",
       "2021-02-10   -1.884436\n",
       "2021-02-11   -1.777037\n",
       "2021-02-12   -1.795237\n",
       "2021-02-15   -1.723338\n",
       "2021-02-16   -1.744338\n",
       "2021-02-17   -1.698336\n",
       "2021-02-18   -1.819638\n",
       "2021-02-19   -1.877537\n",
       "2021-02-22   -1.830536\n",
       "2021-02-23   -1.928937\n",
       "2021-02-24   -1.888136\n",
       "2021-02-25   -2.088337\n",
       "2021-02-26   -1.973338\n",
       "2021-03-01   -1.883338\n",
       "2021-03-02   -1.919337\n",
       "2021-03-03   -1.998337\n",
       "2021-03-04   -2.018338\n",
       "2021-03-05   -1.925337\n",
       "2021-03-08   -1.810837\n",
       "2021-03-09   -1.455336\n",
       "2021-03-10   -1.486336\n",
       "2021-03-11   -1.434336\n",
       "Length: 553, dtype: float64"
      ]
     },
     "execution_count": 134,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 原始值\n",
    "y\n",
    "# 配適值\n",
    "result.fittedvalues\n",
    "# 殘差值\n",
    "result.resid\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}