Chapter3-TimeSeries.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\msharp\\AppData\\Local\\Continuum\\anaconda3\\envs\\ipykernel_py2\\lib\\site-packages\\matplotlib\\backends\\__init__.py:89: MatplotlibDeprecationWarning: matplotlib.verbose is deprecated;\n",
      "Command line argument --verbose-LEVEL is deprecated.\n",
      "This functionality is now provided by the standard\n",
      "python logging library.  To get more (or less) logging output:\n",
      "    import logging\n",
      "    logger = logging.getLogger('matplotlib')\n",
      "    logger.set_level(logging.INFO)\n",
      "  (name, backend_version))\n",
      "C:\\Users\\msharp\\AppData\\Local\\Continuum\\anaconda3\\envs\\ipykernel_py2\\lib\\site-packages\\matplotlib\\__init__.py:356: MatplotlibDeprecationWarning: matplotlib.verbose is deprecated;\n",
      "Command line argument --verbose-LEVEL is deprecated.\n",
      "This functionality is now provided by the standard\n",
      "python logging library.  To get more (or less) logging output:\n",
      "    import logging\n",
      "    logger = logging.getLogger('matplotlib')\n",
      "    logger.set_level(logging.INFO)\n",
      "  @cbook.deprecated(\"2.1\")\n"
     ]
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from datetime import datetime\n",
    "import statsmodels.api as sm\n",
    "import matplotlib.pyplot as plt\n",
    "from IPython.display import Image\n",
    "from matplotlib.pylab import rcParams\n",
    "from statsmodels.tsa.stattools import adfuller\n",
    "from statsmodels.tsa.stattools import acf, pacf\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from statsmodels.tsa.arima_model import ARMA, ARIMA\n",
    "from sklearn.metrics import explained_variance_score\n",
    "from matplotlib.finance import fetch_historical_yahoo, parse_yahoo_historical_ochl\n",
    "# import mpl_finance\n",
    "# from mpl_finance import fetch_historical_yahoo, parse_yahoo_historical_och\n",
    "\n",
    "rcParams['figure.figsize'] = 15, 5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "ename": "HTTPError",
     "evalue": "HTTP Error 502: Bad Gateway",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m\u001b[0m",
      "\u001b[1;31mHTTPError\u001b[0mTraceback (most recent call last)",
      "\u001b[1;32m<ipython-input-8-ec47f6919031>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m()\u001b[0m\n\u001b[0;32m      3\u001b[0m \u001b[0mos\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0menviron\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;34m\"http_proxy\"\u001b[0m\u001b[1;33m]\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;34m'http://proxy-web.micron.com:80'\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      4\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m----> 5\u001b[1;33m \u001b[0mfh\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mfetch_historical_yahoo\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m'^GSPC'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m(\u001b[0m\u001b[1;36m2016\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;36m1\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;36m1\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m(\u001b[0m\u001b[1;36m2016\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;36m12\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;36m31\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      6\u001b[0m \u001b[0myahoo_data\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mparse_yahoo_historical_ochl\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mfh\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0masobject\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mTrue\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0madjusted\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mFalse\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mC:\\Users\\msharp\\AppData\\Local\\Continuum\\anaconda3\\envs\\ipykernel_py2\\lib\\site-packages\\matplotlib\\finance.pyc\u001b[0m in \u001b[0;36mfetch_historical_yahoo\u001b[1;34m(ticker, date1, date2, cachename, dividends)\u001b[0m\n\u001b[0;32m    358\u001b[0m         \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    359\u001b[0m             \u001b[0mmkdirs\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mos\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mpath\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mabspath\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mos\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mpath\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mdirname\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mcachename\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 360\u001b[1;33m             \u001b[1;32mwith\u001b[0m \u001b[0mcontextlib\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mclosing\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0murlopen\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0murl\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;32mas\u001b[0m \u001b[0murlfh\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    361\u001b[0m                 \u001b[1;32mwith\u001b[0m \u001b[0mopen\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mcachename\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m'wb'\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;32mas\u001b[0m \u001b[0mfh\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    362\u001b[0m                     \u001b[0mfh\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mwrite\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0murlfh\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mread\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mC:\\Users\\msharp\\AppData\\Local\\Continuum\\anaconda3\\envs\\ipykernel_py2\\lib\\urllib2.pyc\u001b[0m in \u001b[0;36murlopen\u001b[1;34m(url, data, timeout, cafile, capath, cadefault, context)\u001b[0m\n\u001b[0;32m    152\u001b[0m     \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    153\u001b[0m         \u001b[0mopener\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0m_opener\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 154\u001b[1;33m     \u001b[1;32mreturn\u001b[0m \u001b[0mopener\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mopen\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0murl\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mtimeout\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    155\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    156\u001b[0m \u001b[1;32mdef\u001b[0m \u001b[0minstall_opener\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mopener\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mC:\\Users\\msharp\\AppData\\Local\\Continuum\\anaconda3\\envs\\ipykernel_py2\\lib\\urllib2.pyc\u001b[0m in \u001b[0;36mopen\u001b[1;34m(self, fullurl, data, timeout)\u001b[0m\n\u001b[0;32m    433\u001b[0m         \u001b[1;32mfor\u001b[0m \u001b[0mprocessor\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mprocess_response\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mprotocol\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m[\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    434\u001b[0m             \u001b[0mmeth\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mgetattr\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mprocessor\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmeth_name\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 435\u001b[1;33m             \u001b[0mresponse\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mmeth\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mreq\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mresponse\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    436\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    437\u001b[0m         \u001b[1;32mreturn\u001b[0m \u001b[0mresponse\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mC:\\Users\\msharp\\AppData\\Local\\Continuum\\anaconda3\\envs\\ipykernel_py2\\lib\\urllib2.pyc\u001b[0m in \u001b[0;36mhttp_response\u001b[1;34m(self, request, response)\u001b[0m\n\u001b[0;32m    546\u001b[0m         \u001b[1;32mif\u001b[0m \u001b[1;32mnot\u001b[0m \u001b[1;33m(\u001b[0m\u001b[1;36m200\u001b[0m \u001b[1;33m<=\u001b[0m \u001b[0mcode\u001b[0m \u001b[1;33m<\u001b[0m \u001b[1;36m300\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    547\u001b[0m             response = self.parent.error(\n\u001b[1;32m--> 548\u001b[1;33m                 'http', request, response, code, msg, hdrs)\n\u001b[0m\u001b[0;32m    549\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    550\u001b[0m         \u001b[1;32mreturn\u001b[0m \u001b[0mresponse\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mC:\\Users\\msharp\\AppData\\Local\\Continuum\\anaconda3\\envs\\ipykernel_py2\\lib\\urllib2.pyc\u001b[0m in \u001b[0;36merror\u001b[1;34m(self, proto, *args)\u001b[0m\n\u001b[0;32m    471\u001b[0m         \u001b[1;32mif\u001b[0m \u001b[0mhttp_err\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    472\u001b[0m             \u001b[0margs\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;33m(\u001b[0m\u001b[0mdict\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m'default'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m'http_error_default'\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;33m+\u001b[0m \u001b[0morig_args\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 473\u001b[1;33m             \u001b[1;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_call_chain\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    474\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    475\u001b[0m \u001b[1;31m# XXX probably also want an abstract factory that knows when it makes\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mC:\\Users\\msharp\\AppData\\Local\\Continuum\\anaconda3\\envs\\ipykernel_py2\\lib\\urllib2.pyc\u001b[0m in \u001b[0;36m_call_chain\u001b[1;34m(self, chain, kind, meth_name, *args)\u001b[0m\n\u001b[0;32m    405\u001b[0m             \u001b[0mfunc\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mgetattr\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mhandler\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmeth_name\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    406\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 407\u001b[1;33m             \u001b[0mresult\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mfunc\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    408\u001b[0m             \u001b[1;32mif\u001b[0m \u001b[0mresult\u001b[0m \u001b[1;32mis\u001b[0m \u001b[1;32mnot\u001b[0m \u001b[0mNone\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    409\u001b[0m                 \u001b[1;32mreturn\u001b[0m \u001b[0mresult\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mC:\\Users\\msharp\\AppData\\Local\\Continuum\\anaconda3\\envs\\ipykernel_py2\\lib\\urllib2.pyc\u001b[0m in \u001b[0;36mhttp_error_default\u001b[1;34m(self, req, fp, code, msg, hdrs)\u001b[0m\n\u001b[0;32m    554\u001b[0m \u001b[1;32mclass\u001b[0m \u001b[0mHTTPDefaultErrorHandler\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mBaseHandler\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    555\u001b[0m     \u001b[1;32mdef\u001b[0m \u001b[0mhttp_error_default\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mreq\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mfp\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mcode\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmsg\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mhdrs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 556\u001b[1;33m         \u001b[1;32mraise\u001b[0m \u001b[0mHTTPError\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mreq\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget_full_url\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mcode\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmsg\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mhdrs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mfp\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    557\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    558\u001b[0m \u001b[1;32mclass\u001b[0m \u001b[0mHTTPRedirectHandler\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mBaseHandler\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mHTTPError\u001b[0m: HTTP Error 502: Bad Gateway"
     ]
    }
   ],
   "source": [
    "import os\n",
    "os.environ[\"https_proxy\"] = 'http://proxy-web.micron.com:80'\n",
    "os.environ[\"http_proxy\"] = 'http://proxy-web.micron.com:80'\n",
    "\n",
    "fh = fetch_historical_yahoo('^GSPC', (2016, 1, 1), (2016, 12, 31))\n",
    "yahoo_data = parse_yahoo_historical_ochl(fh, asobject=True, adjusted=False)\n",
    "# after much investigation looks like yahoo API is discontinued sometime around May 2017\n",
    "# this book was published sometime 2017 (probably earlier)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "dates = [x[0] for x in list(yahoo_data)]\n",
    "values = [x[-1] for x in list(yahoo_data)]\n",
    "data = pd.DataFrame({'Close_Adj':values}, index=dates)\n",
    "\n",
    "date_thresh = datetime.strptime('2016-10-01',\"%Y-%m-%d\").date()\n",
    "\n",
    "data_train = data[:date_thresh]\n",
    "data_test = data[date_thresh:]\n",
    "\n",
    "plt.plot(data_train.index, data['Close_Adj'])\n",
    "plt.xlabel('Year')\n",
    "plt.ylabel('Adjusted Closing Rate')\n",
    "plt.title('Yahoo Adjusted closing rate - Year 2016')"
   ]
  },
  {
   "attachments": {
    "image.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "![image.png](attachment:image.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def evaluate_stationarity(timeseries, t=30):\n",
    "    \n",
    "    #Determing rolling statistics\n",
    "    rolmean = timeseries.rolling(window=t).mean()\n",
    "    rolstd = timeseries.rolling(window=t).std()\n",
    "\n",
    "    #Plot rolling statistics:\n",
    "    orig = plt.plot(timeseries, color='blue',label='Original')\n",
    "    mean = plt.plot(rolmean, color='red', label='Rolling Mean')\n",
    "    std = plt.plot(rolstd, color='black', label = 'Rolling Std')\n",
    "    plt.legend(loc='best')\n",
    "    plt.title('Rolling Mean & Standard Deviation')\n",
    "    plt.show(block=False)\n",
    "    \n",
    "    #Perform Dickey-Fuller test:\n",
    "    print 'Results of Dickey-Fuller Test:'\n",
    "    dftest = adfuller(timeseries, autolag='AIC')\n",
    "    dfoutput = pd.Series(dftest[0:4], index=['Test Statistic','p-value','#Lags Used','Number of Observations Used'])\n",
    "    for key,value in dftest[4].items():\n",
    "        dfoutput['Critical Value (%s)'%key] = value\n",
    "    print dfoutput"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "evaluate_stationarity(data_train['Close_Adj'], 15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_log = np.log(data_train['Close_Adj'])\n",
    "evaluate_stationarity(data_log, 15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_log = np.log(data_train['Close_Adj'])\n",
    "\n",
    "#Determing rolling statistics\n",
    "rolmean = data_log.rolling(window=15).mean()\n",
    "rolstd = data_log.rolling(window=15).std()\n",
    "\n",
    "#Plot rolling statistics:\n",
    "orig = plt.plot(data_log, color='blue',label='Original')\n",
    "mean = plt.plot(rolmean, color='red', label='Rolling Mean')\n",
    "std = plt.plot(rolstd, color='black', label = 'Rolling Std')\n",
    "plt.legend(loc='best')\n",
    "plt.ylim([7.4,7.8])\n",
    "plt.title('Rolling Mean & Standard Deviation')\n",
    "plt.show(block=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_sqrt = np.sqrt(data_train['Close_Adj'])\n",
    "evaluate_stationarity(data_sqrt, 15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "moving_avg = pd.rolling_mean(data_log,15)\n",
    "plt.plot(data_log)\n",
    "plt.plot(moving_avg, color='red')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_log_moving_avg_diff = data_log - moving_avg\n",
    "data_log_moving_avg_diff.head(15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_log_moving_avg_diff.dropna(inplace=True)\n",
    "evaluate_stationarity(data_log_moving_avg_diff)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "expwighted_avg = pd.ewma(data_log, halflife=15)\n",
    "data_log_ewma_diff = data_log - expwighted_avg\n",
    "evaluate_stationarity(data_log_ewma_diff)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_log_diff = data_log - data_log.shift()\n",
    "data_log_diff.dropna(inplace=True)\n",
    "evaluate_stationarity(data_log_diff)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Excercise: Apply second- and third-order differencing to our log transformed series. Did it make the log transformed series stationary?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Decomposition - breaking to trend, seasonality, and residuals\n",
    "from statsmodels.tsa.seasonal import seasonal_decompose\n",
    "decomposition = seasonal_decompose(list(data_log), freq=15)\n",
    "\n",
    "trend = decomposition.trend\n",
    "seasonal = decomposition.seasonal\n",
    "residual = decomposition.resid\n",
    "\n",
    "plt.subplot(411)\n",
    "plt.plot(data_log, label='Original')\n",
    "plt.legend(loc='best')\n",
    "plt.subplot(412)\n",
    "plt.plot(trend, label='Trend')\n",
    "plt.legend(loc='best')\n",
    "plt.subplot(413)\n",
    "plt.plot(seasonal,label='Seasonality')\n",
    "plt.legend(loc='best')\n",
    "plt.subplot(414)\n",
    "plt.plot(residual, label='Residuals')\n",
    "plt.legend(loc='best')\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_log_decompose = pd.Series(residual)\n",
    "data_log_decompose.dropna(inplace=True)\n",
    "evaluate_stationarity(data_log_decompose)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Durbin Watson Statistic\n",
    "Contrary to the autocorrelation correlogram, Durbin Watson statistic is more of a statistical approach to determining the existence of autocorrelation within the data. It does so by computing autocorrelation among residuals from a statistical regression analysis. It returns a number between 0 and 4 where 0 depicts strong positive autocorrelation, 4 depicts strong negative autocorrelation, and 2 depicts no autocorrelation at all."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sm.stats.durbin_watson(data_log_diff)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ax1 = plt.subplot(211)\n",
    "fig = sm.graphics.tsa.plot_acf(data_log_diff.squeeze(), lags=40, ax=ax1)\n",
    "ax2 = plt.subplot(212)\n",
    "fig = sm.graphics.tsa.plot_pacf(data_log_diff, lags=40, ax=ax2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# ARIMA(p,d,q) - Auto-Regressive Integrated Moving Averages\n",
    "# AR - Auto-Regressive -> q=0\n",
    "model = ARIMA(data_log, order=(2, 1, 0), dates=data_log.index.to_datetime())\n",
    "results_AR_210 = model.fit(disp=-1)\n",
    "plt.plot(data_log_diff)\n",
    "plt.plot(results_AR_210.fittedvalues, color='red')\n",
    "plt.title('RSS: %.4f'% sum((results_AR_210.fittedvalues-data_log_diff)**2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# MA - Moving Average -> p=0\n",
    "model = ARIMA(data_log, order=(0, 1, 1), dates=data_log.index.to_datetime())\n",
    "results_MA_011 = model.fit(disp=-1)\n",
    "plt.plot(data_log_diff)\n",
    "plt.plot(results_MA_011.fittedvalues, color='red')\n",
    "plt.title('RSS: %.4f'% sum((results_MA_011.fittedvalues-data_log_diff)**2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = ARIMA(data_log, order=(1, 1, 1), dates=data_log.index.to_datetime())\n",
    "results_ARIMA_111 = model.fit(disp=-1)\n",
    "plt.plot(data_log_diff)\n",
    "plt.plot(results_ARIMA_111.fittedvalues, color='red')\n",
    "plt.title('RSS: %.4f'% sum((results_ARIMA_111.fittedvalues-data_log_diff)**2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions_ARIMA_diff = pd.Series(results_ARIMA_111.fittedvalues,\n",
    "copy=True)\n",
    "print predictions_ARIMA_diff.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions_ARIMA_diff_cumsum = predictions_ARIMA_diff.cumsum()\n",
    "print predictions_ARIMA_diff_cumsum.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print results_ARIMA_111.aic"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "predictions_ARIMA_log = pd.Series(data_log.ix[0], index=data_log.index)\n",
    "predictions_ARIMA_log = predictions_ARIMA_log.add(predictions_ARIMA_diff_\n",
    "cumsum,fill_value=0)\n",
    "predictions_ARIMA_log.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def mean_forecast_err(y, yhat):\n",
    "    return y.sub(yhat).mean()\n",
    "\n",
    "def mean_absolute_err(y, yhat):\n",
    "    return np.mean((np.abs(y.sub(yhat).mean()) / yhat)) # or\n",
    "    percent error = * 100\n",
    "\n",
    "def rmse(y, yhat):\n",
    "    return np.sqrt(sum((yhat-y)**2)/len(y))\n",
    "\n",
    "predictions_ARIMA = np.exp(predictions_ARIMA_log)\n",
    "plt.plot(data_train['Close_Adj'])\n",
    "plt.plot(predictions_ARIMA)\n",
    "\n",
    "plt.title('RMSE: %.4f | MFE: %.4f | MAE: %.4f'% (\n",
    "        rmse(data_train['Close_Adj'], predictions_ARIMA), mean_forecast_\n",
    "            err(data_train['Close_Adj'], predictions_ARIMA), mean_absolute_\n",
    "            err(data_train['Close_Adj'], predictions_ARIMA)))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "regr = LinearRegression()\n",
    "\n",
    "x_train = [[x] for x in range(len(data_train))]\n",
    "y_train = [[y] for y in list(data_train['Close_Adj'])]\n",
    "\n",
    "x_test = [[z] for z in range(x+1, x + 1+len(data_test))]\n",
    "\n",
    "regr.fit(x_train, y_train)\n",
    "\n",
    "y_pred = regr.predict(x_test)\n",
    "\n",
    "explained_variance_score(data_test['Close_Adj'], y_pred)\n",
    "\n",
    "plt.scatter(range(len(data_test)), data_test['Close_Adj'], color='black')\n",
    "plt.plot(range(len(data_test)), y_pred, color='blue',\n",
    "             linewidth=3)\n",
    "plt.ylabel('Concrete strength')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.15"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}