module4/pca_basics.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# PCA Basics - Coding Exercises\n",
    "\n",
    "In this notebook you will apply what you learned about PCA in the previous lesson to real stock data."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Install Packages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting zipline===1.3.0 (from -r requirements.txt (line 1))\n",
      "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/be/59/8c5802a7897c1095fdc409fb557f04df8f75c37174e80d2ba58c8d8a6488/zipline-1.3.0.tar.gz (2.5MB)\n",
      "\u001b[K    100% |████████████████████████████████| 2.5MB 8.6MB/s eta 0:00:01\n",
      "\u001b[?25hRequirement already satisfied: pip>=7.1.0 in /opt/conda/lib/python3.6/site-packages (from zipline===1.3.0->-r requirements.txt (line 1)) (18.1)\n",
      "Requirement already satisfied: setuptools>18.0 in /opt/conda/lib/python3.6/site-packages (from zipline===1.3.0->-r requirements.txt (line 1)) (38.4.0)\n",
      "Collecting Logbook>=0.12.5 (from zipline===1.3.0->-r requirements.txt (line 1))\n",
      "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/2f/d9/16ac346f7c0102835814cc9e5b684aaadea101560bb932a2403bd26b2320/Logbook-1.5.3.tar.gz (85kB)\n",
      "\u001b[K    100% |████████████████████████████████| 92kB 16.2MB/s ta 0:00:01\n",
      "\u001b[?25hRequirement already satisfied: pytz>=2016.4 in /opt/conda/lib/python3.6/site-packages (from zipline===1.3.0->-r requirements.txt (line 1)) (2017.3)\n",
      "Requirement already satisfied: numpy>=1.11.1 in /opt/conda/lib/python3.6/site-packages (from zipline===1.3.0->-r requirements.txt (line 1)) (1.12.1)\n",
      "Collecting requests-file>=1.4.1 (from zipline===1.3.0->-r requirements.txt (line 1))\n",
      "  Downloading https://files.pythonhosted.org/packages/23/9c/6e63c23c39e53d3df41c77a3d05a49a42c4e1383a6d2a5e3233161b89dbf/requests_file-1.4.3-py2.py3-none-any.whl\n",
      "Requirement already satisfied: scipy>=0.17.1 in /opt/conda/lib/python3.6/site-packages (from zipline===1.3.0->-r requirements.txt (line 1)) (0.19.1)\n",
      "Collecting pandas<=0.22,>=0.18.1 (from zipline===1.3.0->-r requirements.txt (line 1))\n",
      "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/da/c6/0936bc5814b429fddb5d6252566fe73a3e40372e6ceaf87de3dec1326f28/pandas-0.22.0-cp36-cp36m-manylinux1_x86_64.whl (26.2MB)\n",
      "\u001b[K    100% |████████████████████████████████| 26.3MB 1.3MB/s eta 0:00:01  2% |▉                               | 665kB 33.2MB/s eta 0:00:01    8% |██▉                             | 2.4MB 32.7MB/s eta 0:00:01    21% |██████▉                         | 5.6MB 33.7MB/s eta 0:00:01    27% |████████▊                       | 7.2MB 31.5MB/s eta 0:00:01    45% |██████████████▌                 | 11.9MB 32.8MB/s eta 0:00:01    51% |████████████████▍               | 13.5MB 30.9MB/s eta 0:00:01    74% |████████████████████████        | 19.6MB 32.8MB/s eta 0:00:01    80% |█████████████████████████▊      | 21.1MB 32.7MB/s eta 0:00:01    86% |███████████████████████████▋    | 22.6MB 29.6MB/s eta 0:00:01\n",
      "\u001b[?25hCollecting pandas-datareader>=0.2.1 (from zipline===1.3.0->-r requirements.txt (line 1))\n",
      "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/14/52/accb990baebe0063977f26e02df36aa7eb4015ed4e86f828cd76273cd6f1/pandas_datareader-0.8.1-py2.py3-none-any.whl (107kB)\n",
      "\u001b[K    100% |████████████████████████████████| 112kB 19.3MB/s ta 0:00:01\n",
      "\u001b[?25hRequirement already satisfied: patsy>=0.4.0 in /opt/conda/lib/python3.6/site-packages (from zipline===1.3.0->-r requirements.txt (line 1)) (0.4.1)\n",
      "Requirement already satisfied: statsmodels>=0.6.1 in /opt/conda/lib/python3.6/site-packages (from zipline===1.3.0->-r requirements.txt (line 1)) (0.8.0)\n",
      "Requirement already satisfied: python-dateutil>=2.4.2 in /opt/conda/lib/python3.6/site-packages (from zipline===1.3.0->-r requirements.txt (line 1)) (2.6.1)\n",
      "Requirement already satisfied: six>=1.10.0 in /opt/conda/lib/python3.6/site-packages (from zipline===1.3.0->-r requirements.txt (line 1)) (1.11.0)\n",
      "Requirement already satisfied: requests>=2.9.1 in /opt/conda/lib/python3.6/site-packages (from zipline===1.3.0->-r requirements.txt (line 1)) (2.18.4)\n",
      "Requirement already satisfied: Cython>=0.25.2 in /opt/conda/lib/python3.6/site-packages (from zipline===1.3.0->-r requirements.txt (line 1)) (0.29.7)\n",
      "Collecting cyordereddict>=0.2.2 (from zipline===1.3.0->-r requirements.txt (line 1))\n",
      "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/d1/1a/364cbfd927be1b743c7f0a985a7f1f7e8a51469619f9fefe4ee9240ba210/cyordereddict-1.0.0.tar.gz (138kB)\n",
      "\u001b[K    100% |████████████████████████████████| 143kB 21.8MB/s ta 0:00:01\n",
      "\u001b[?25hCollecting bottleneck>=1.0.0 (from zipline===1.3.0->-r requirements.txt (line 1))\n",
      "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/05/ae/cedf5323f398ab4e4ff92d6c431a3e1c6a186f9b41ab3e8258dff786a290/Bottleneck-1.2.1.tar.gz (105kB)\n",
      "\u001b[K    100% |████████████████████████████████| 112kB 20.2MB/s ta 0:00:01\n",
      "\u001b[?25hCollecting contextlib2>=0.4.0 (from zipline===1.3.0->-r requirements.txt (line 1))\n",
      "  Downloading https://files.pythonhosted.org/packages/85/60/370352f7ef6aa96c52fb001831622f50f923c1d575427d021b8ab3311236/contextlib2-0.6.0.post1-py2.py3-none-any.whl\n",
      "Requirement already satisfied: decorator>=4.0.0 in /opt/conda/lib/python3.6/site-packages (from zipline===1.3.0->-r requirements.txt (line 1)) (4.0.11)\n",
      "Requirement already satisfied: networkx<2.0,>=1.9.1 in /opt/conda/lib/python3.6/site-packages (from zipline===1.3.0->-r requirements.txt (line 1)) (1.11)\n",
      "Requirement already satisfied: numexpr>=2.6.1 in /opt/conda/lib/python3.6/site-packages (from zipline===1.3.0->-r requirements.txt (line 1)) (2.6.4)\n",
      "Collecting bcolz<1,>=0.12.1 (from zipline===1.3.0->-r requirements.txt (line 1))\n",
      "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/6c/8b/1ffa01f872cac36173c5eb95b58c01040d8d25f1b242c48577f4104cd3ab/bcolz-0.12.1.tar.gz (622kB)\n",
      "\u001b[K    100% |████████████████████████████████| 624kB 16.2MB/s ta 0:00:01\n",
      "\u001b[?25hRequirement already satisfied: click>=4.0.0 in /opt/conda/lib/python3.6/site-packages (from zipline===1.3.0->-r requirements.txt (line 1)) (6.7)\n",
      "Requirement already satisfied: toolz>=0.8.2 in /opt/conda/lib/python3.6/site-packages (from zipline===1.3.0->-r requirements.txt (line 1)) (0.8.2)\n",
      "Collecting multipledispatch>=0.4.8 (from zipline===1.3.0->-r requirements.txt (line 1))\n",
      "  Downloading https://files.pythonhosted.org/packages/89/79/429ecef45fd5e4504f7474d4c3c3c4668c267be3370e4c2fd33e61506833/multipledispatch-0.6.0-py3-none-any.whl\n",
      "Requirement already satisfied: MarkupSafe>=0.23 in /opt/conda/lib/python3.6/site-packages (from zipline===1.3.0->-r requirements.txt (line 1)) (1.0)\n",
      "Requirement already satisfied: Mako>=1.0.1 in /opt/conda/lib/python3.6/site-packages/Mako-1.0.7-py3.6.egg (from zipline===1.3.0->-r requirements.txt (line 1)) (1.0.7)\n",
      "Requirement already satisfied: sqlalchemy>=1.0.8 in /opt/conda/lib/python3.6/site-packages (from zipline===1.3.0->-r requirements.txt (line 1)) (1.1.13)\n",
      "Collecting alembic>=0.7.7 (from zipline===1.3.0->-r requirements.txt (line 1))\n",
      "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/70/3d/d5ed7a71fe84f9ed0a69e91232a40b0b148b151524dc5bb1c8e4211eb117/alembic-1.3.0.tar.gz (1.1MB)\n",
      "\u001b[K    100% |████████████████████████████████| 1.1MB 13.9MB/s ta 0:00:01\n",
      "\u001b[?25hCollecting sortedcontainers>=1.4.4 (from zipline===1.3.0->-r requirements.txt (line 1))\n",
      "  Downloading https://files.pythonhosted.org/packages/13/f3/cf85f7c3a2dbd1a515d51e1f1676d971abe41bba6f4ab5443240d9a78e5b/sortedcontainers-2.1.0-py2.py3-none-any.whl\n",
      "Collecting intervaltree>=2.1.0 (from zipline===1.3.0->-r requirements.txt (line 1))\n",
      "  Downloading https://files.pythonhosted.org/packages/e8/f9/76237755b2020cd74549e98667210b2dd54d3fb17c6f4a62631e61d31225/intervaltree-3.0.2.tar.gz\n",
      "Collecting lru-dict>=1.1.4 (from zipline===1.3.0->-r requirements.txt (line 1))\n",
      "  Downloading https://files.pythonhosted.org/packages/00/a5/32ed6e10246cd341ca8cc205acea5d208e4053f48a4dced2b1b31d45ba3f/lru-dict-1.1.6.tar.gz\n",
      "Collecting empyrical>=0.5.0 (from zipline===1.3.0->-r requirements.txt (line 1))\n",
      "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/84/9e/9506e8b25464ff57ef93b5ba9092b464b44dc76b717695b126b3c93214a2/empyrical-0.5.3.tar.gz (50kB)\n",
      "\u001b[K    100% |████████████████████████████████| 51kB 13.0MB/s ta 0:00:01\n",
      "\u001b[?25hCollecting tables>=3.3.0 (from zipline===1.3.0->-r requirements.txt (line 1))\n",
      "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/ed/c3/8fd9e3bb21872f9d69eb93b3014c86479864cca94e625fd03713ccacec80/tables-3.6.1-cp36-cp36m-manylinux1_x86_64.whl (4.3MB)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[K    100% |████████████████████████████████| 4.3MB 8.0MB/s eta 0:00:01    60% |███████████████████▎            | 2.6MB 28.3MB/s eta 0:00:01    91% |█████████████████████████████▎  | 4.0MB 28.8MB/s eta 0:00:01\n",
      "\u001b[?25hCollecting trading-calendars>=1.0.1 (from zipline===1.3.0->-r requirements.txt (line 1))\n",
      "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/41/f8/f0b074db66485a8127ca7b5fbdc5aa4eb93a810113266384aac59d6e1ff6/trading_calendars-1.11.0.tar.gz (96kB)\n",
      "\u001b[K    100% |████████████████████████████████| 102kB 16.1MB/s a 0:00:01\n",
      "\u001b[?25hRequirement already satisfied: lxml in /opt/conda/lib/python3.6/site-packages (from pandas-datareader>=0.2.1->zipline===1.3.0->-r requirements.txt (line 1)) (4.1.1)\n",
      "Requirement already satisfied: chardet<3.1.0,>=3.0.2 in /opt/conda/lib/python3.6/site-packages (from requests>=2.9.1->zipline===1.3.0->-r requirements.txt (line 1)) (3.0.4)\n",
      "Requirement already satisfied: idna<2.7,>=2.5 in /opt/conda/lib/python3.6/site-packages (from requests>=2.9.1->zipline===1.3.0->-r requirements.txt (line 1)) (2.6)\n",
      "Requirement already satisfied: urllib3<1.23,>=1.21.1 in /opt/conda/lib/python3.6/site-packages (from requests>=2.9.1->zipline===1.3.0->-r requirements.txt (line 1)) (1.22)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in /opt/conda/lib/python3.6/site-packages (from requests>=2.9.1->zipline===1.3.0->-r requirements.txt (line 1)) (2017.11.5)\n",
      "Collecting python-editor>=0.3 (from alembic>=0.7.7->zipline===1.3.0->-r requirements.txt (line 1))\n",
      "  Downloading https://files.pythonhosted.org/packages/c6/d3/201fc3abe391bbae6606e6f1d598c15d367033332bd54352b12f35513717/python_editor-1.0.4-py3-none-any.whl\n",
      "Building wheels for collected packages: zipline, Logbook, cyordereddict, bottleneck, bcolz, alembic, intervaltree, lru-dict, empyrical, trading-calendars\n",
      "  Running setup.py bdist_wheel for zipline ... \u001b[?25ldone\n",
      "\u001b[?25h  Stored in directory: /root/.cache/pip/wheels/a4/d6/67/f303ab028b004bf8e00c05b5b04fba83d8ec238b6547becdb7\n",
      "  Running setup.py bdist_wheel for Logbook ... \u001b[?25ldone\n",
      "\u001b[?25h  Stored in directory: /root/.cache/pip/wheels/d2/70/07/68b99a8e05dcd1ab194a8e0ccb9e4d0ac5dd6d8d139c7149b4\n",
      "  Running setup.py bdist_wheel for cyordereddict ... \u001b[?25ldone\n",
      "\u001b[?25h  Stored in directory: /root/.cache/pip/wheels/0b/9d/8b/5bf3e22c1edd59b50f11bb19dec9dfcfe5a479fc7ace02b61f\n",
      "  Running setup.py bdist_wheel for bottleneck ... \u001b[?25ldone\n",
      "\u001b[?25h  Stored in directory: /root/.cache/pip/wheels/f2/bf/ec/e0f39aa27001525ad455139ee57ec7d0776fe074dfd78c97e4\n",
      "  Running setup.py bdist_wheel for bcolz ... \u001b[?25ldone\n",
      "\u001b[?25h  Stored in directory: /root/.cache/pip/wheels/c5/cc/1b/2cf1f88959af5d7f4d449b7fc6c9452d0ecbd86fd61a9ee376\n",
      "  Running setup.py bdist_wheel for alembic ... \u001b[?25ldone\n",
      "\u001b[?25h  Stored in directory: /root/.cache/pip/wheels/40/f8/22/ad0f408796a4c656fae5ee1fd8d8a139b19ca4af61059cea5b\n",
      "  Running setup.py bdist_wheel for intervaltree ... \u001b[?25ldone\n",
      "\u001b[?25h  Stored in directory: /root/.cache/pip/wheels/08/99/c0/5a5942f5b9567c59c14aac76f95a70bf11dccc71240b91ebf5\n",
      "  Running setup.py bdist_wheel for lru-dict ... \u001b[?25ldone\n",
      "\u001b[?25h  Stored in directory: /root/.cache/pip/wheels/b7/ef/06/fbdd555907a7d438fb33e4c8675f771ff1cf41917284c51ebf\n",
      "  Running setup.py bdist_wheel for empyrical ... \u001b[?25ldone\n",
      "\u001b[?25h  Stored in directory: /root/.cache/pip/wheels/10/a4/3b/951bd609878a82fd72b9ea23699daf1eaada4ff6f583152876\n",
      "  Running setup.py bdist_wheel for trading-calendars ... \u001b[?25ldone\n",
      "\u001b[?25h  Stored in directory: /root/.cache/pip/wheels/e4/7e/55/10fc81c7c90d3f2d0e9e35d08b21481a6e335751c412712b9b\n",
      "Successfully built zipline Logbook cyordereddict bottleneck bcolz alembic intervaltree lru-dict empyrical trading-calendars\n",
      "Installing collected packages: Logbook, requests-file, pandas, pandas-datareader, cyordereddict, bottleneck, contextlib2, bcolz, multipledispatch, python-editor, alembic, sortedcontainers, intervaltree, lru-dict, empyrical, tables, trading-calendars, zipline\n",
      "  Found existing installation: pandas 0.23.3\n",
      "    Uninstalling pandas-0.23.3:\n",
      "      Successfully uninstalled pandas-0.23.3\n",
      "Successfully installed Logbook-1.5.3 alembic-1.3.0 bcolz-0.12.1 bottleneck-1.2.1 contextlib2-0.6.0.post1 cyordereddict-1.0.0 empyrical-0.5.3 intervaltree-3.0.2 lru-dict-1.1.6 multipledispatch-0.6.0 pandas-0.22.0 pandas-datareader-0.8.1 python-editor-1.0.4 requests-file-1.4.3 sortedcontainers-2.1.0 tables-3.6.1 trading-calendars-1.11.0 zipline-1.3.0\n"
     ]
    }
   ],
   "source": [
    "import sys\n",
    "!{sys.executable} -m pip install -r requirements.txt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Get Returns\n",
    "\n",
    "In the previous lesson we used 2-dimensional randomly correlated data to see how we can use PCA to for dimensionality reduction. In this notebook, we will do the same but for 490-dimensional data of stock returns. We will get the stock returns using Zipline and data from Quotemedia, just as we learned in previous lessons. The function `get_returns(start_date, end_date)` in the `utils` module, gets the data from the Quotemedia data bundle and produces the stock returns for the given `start_date` and `end_date`. You are welcome to take a look at the `utils` module to see how this is done.\n",
    "\n",
    "In the code below, we use `utils.get_returns` funtion to get the returns for stock data between `2011-01-05` and `2016-01-05`. You can change the start and end dates, but if you do, you have to make sure the dates are valid trading dates. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>Equity(0 [A])</th>\n",
       "      <th>Equity(1 [AAL])</th>\n",
       "      <th>Equity(2 [AAP])</th>\n",
       "      <th>Equity(3 [AAPL])</th>\n",
       "      <th>Equity(4 [ABBV])</th>\n",
       "      <th>Equity(5 [ABC])</th>\n",
       "      <th>Equity(6 [ABT])</th>\n",
       "      <th>Equity(7 [ACN])</th>\n",
       "      <th>Equity(8 [ADBE])</th>\n",
       "      <th>Equity(9 [ADI])</th>\n",
       "      <th>...</th>\n",
       "      <th>Equity(481 [XL])</th>\n",
       "      <th>Equity(482 [XLNX])</th>\n",
       "      <th>Equity(483 [XOM])</th>\n",
       "      <th>Equity(484 [XRAY])</th>\n",
       "      <th>Equity(485 [XRX])</th>\n",
       "      <th>Equity(486 [XYL])</th>\n",
       "      <th>Equity(487 [YUM])</th>\n",
       "      <th>Equity(488 [ZBH])</th>\n",
       "      <th>Equity(489 [ZION])</th>\n",
       "      <th>Equity(490 [ZTS])</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2011-01-07 00:00:00+00:00</th>\n",
       "      <td>0.008437</td>\n",
       "      <td>0.014230</td>\n",
       "      <td>0.026702</td>\n",
       "      <td>0.007146</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.001994</td>\n",
       "      <td>0.004165</td>\n",
       "      <td>0.001648</td>\n",
       "      <td>-0.007127</td>\n",
       "      <td>-0.005818</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.001838</td>\n",
       "      <td>-0.005619</td>\n",
       "      <td>0.005461</td>\n",
       "      <td>-0.004044</td>\n",
       "      <td>-0.013953</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.012457</td>\n",
       "      <td>-0.000181</td>\n",
       "      <td>-0.010458</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-01-10 00:00:00+00:00</th>\n",
       "      <td>-0.004174</td>\n",
       "      <td>0.006195</td>\n",
       "      <td>0.007435</td>\n",
       "      <td>0.018852</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-0.005714</td>\n",
       "      <td>-0.008896</td>\n",
       "      <td>-0.008854</td>\n",
       "      <td>0.028714</td>\n",
       "      <td>0.002926</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000947</td>\n",
       "      <td>0.007814</td>\n",
       "      <td>-0.006081</td>\n",
       "      <td>0.010466</td>\n",
       "      <td>0.009733</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.001440</td>\n",
       "      <td>0.007784</td>\n",
       "      <td>-0.017945</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-01-11 00:00:00+00:00</th>\n",
       "      <td>-0.001886</td>\n",
       "      <td>-0.043644</td>\n",
       "      <td>-0.005927</td>\n",
       "      <td>-0.002367</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.009783</td>\n",
       "      <td>-0.002067</td>\n",
       "      <td>0.013717</td>\n",
       "      <td>0.000607</td>\n",
       "      <td>0.008753</td>\n",
       "      <td>...</td>\n",
       "      <td>0.001314</td>\n",
       "      <td>0.010179</td>\n",
       "      <td>0.007442</td>\n",
       "      <td>0.007351</td>\n",
       "      <td>0.006116</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-0.006470</td>\n",
       "      <td>0.035676</td>\n",
       "      <td>0.007467</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-01-12 00:00:00+00:00</th>\n",
       "      <td>0.017254</td>\n",
       "      <td>-0.008237</td>\n",
       "      <td>0.013387</td>\n",
       "      <td>0.008133</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-0.005979</td>\n",
       "      <td>-0.001011</td>\n",
       "      <td>0.022969</td>\n",
       "      <td>0.017950</td>\n",
       "      <td>0.000257</td>\n",
       "      <td>...</td>\n",
       "      <td>0.004986</td>\n",
       "      <td>0.015666</td>\n",
       "      <td>0.011763</td>\n",
       "      <td>0.027182</td>\n",
       "      <td>0.004386</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.002631</td>\n",
       "      <td>0.014741</td>\n",
       "      <td>-0.011903</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-01-13 00:00:00+00:00</th>\n",
       "      <td>-0.004559</td>\n",
       "      <td>0.000955</td>\n",
       "      <td>0.003031</td>\n",
       "      <td>0.003657</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.014925</td>\n",
       "      <td>-0.004451</td>\n",
       "      <td>-0.000400</td>\n",
       "      <td>-0.005719</td>\n",
       "      <td>-0.005012</td>\n",
       "      <td>...</td>\n",
       "      <td>0.030499</td>\n",
       "      <td>-0.003217</td>\n",
       "      <td>0.001694</td>\n",
       "      <td>0.000547</td>\n",
       "      <td>-0.018235</td>\n",
       "      <td>0.0</td>\n",
       "      <td>-0.005084</td>\n",
       "      <td>-0.004665</td>\n",
       "      <td>-0.009178</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 490 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                           Equity(0 [A])  Equity(1 [AAL])  Equity(2 [AAP])  \\\n",
       "2011-01-07 00:00:00+00:00       0.008437         0.014230         0.026702   \n",
       "2011-01-10 00:00:00+00:00      -0.004174         0.006195         0.007435   \n",
       "2011-01-11 00:00:00+00:00      -0.001886        -0.043644        -0.005927   \n",
       "2011-01-12 00:00:00+00:00       0.017254        -0.008237         0.013387   \n",
       "2011-01-13 00:00:00+00:00      -0.004559         0.000955         0.003031   \n",
       "\n",
       "                           Equity(3 [AAPL])  Equity(4 [ABBV])  \\\n",
       "2011-01-07 00:00:00+00:00          0.007146               0.0   \n",
       "2011-01-10 00:00:00+00:00          0.018852               0.0   \n",
       "2011-01-11 00:00:00+00:00         -0.002367               0.0   \n",
       "2011-01-12 00:00:00+00:00          0.008133               0.0   \n",
       "2011-01-13 00:00:00+00:00          0.003657               0.0   \n",
       "\n",
       "                           Equity(5 [ABC])  Equity(6 [ABT])  Equity(7 [ACN])  \\\n",
       "2011-01-07 00:00:00+00:00         0.001994         0.004165         0.001648   \n",
       "2011-01-10 00:00:00+00:00        -0.005714        -0.008896        -0.008854   \n",
       "2011-01-11 00:00:00+00:00         0.009783        -0.002067         0.013717   \n",
       "2011-01-12 00:00:00+00:00        -0.005979        -0.001011         0.022969   \n",
       "2011-01-13 00:00:00+00:00         0.014925        -0.004451        -0.000400   \n",
       "\n",
       "                           Equity(8 [ADBE])  Equity(9 [ADI])  \\\n",
       "2011-01-07 00:00:00+00:00         -0.007127        -0.005818   \n",
       "2011-01-10 00:00:00+00:00          0.028714         0.002926   \n",
       "2011-01-11 00:00:00+00:00          0.000607         0.008753   \n",
       "2011-01-12 00:00:00+00:00          0.017950         0.000257   \n",
       "2011-01-13 00:00:00+00:00         -0.005719        -0.005012   \n",
       "\n",
       "                                 ...          Equity(481 [XL])  \\\n",
       "2011-01-07 00:00:00+00:00        ...                 -0.001838   \n",
       "2011-01-10 00:00:00+00:00        ...                  0.000947   \n",
       "2011-01-11 00:00:00+00:00        ...                  0.001314   \n",
       "2011-01-12 00:00:00+00:00        ...                  0.004986   \n",
       "2011-01-13 00:00:00+00:00        ...                  0.030499   \n",
       "\n",
       "                           Equity(482 [XLNX])  Equity(483 [XOM])  \\\n",
       "2011-01-07 00:00:00+00:00           -0.005619           0.005461   \n",
       "2011-01-10 00:00:00+00:00            0.007814          -0.006081   \n",
       "2011-01-11 00:00:00+00:00            0.010179           0.007442   \n",
       "2011-01-12 00:00:00+00:00            0.015666           0.011763   \n",
       "2011-01-13 00:00:00+00:00           -0.003217           0.001694   \n",
       "\n",
       "                           Equity(484 [XRAY])  Equity(485 [XRX])  \\\n",
       "2011-01-07 00:00:00+00:00           -0.004044          -0.013953   \n",
       "2011-01-10 00:00:00+00:00            0.010466           0.009733   \n",
       "2011-01-11 00:00:00+00:00            0.007351           0.006116   \n",
       "2011-01-12 00:00:00+00:00            0.027182           0.004386   \n",
       "2011-01-13 00:00:00+00:00            0.000547          -0.018235   \n",
       "\n",
       "                           Equity(486 [XYL])  Equity(487 [YUM])  \\\n",
       "2011-01-07 00:00:00+00:00                0.0           0.012457   \n",
       "2011-01-10 00:00:00+00:00                0.0           0.001440   \n",
       "2011-01-11 00:00:00+00:00                0.0          -0.006470   \n",
       "2011-01-12 00:00:00+00:00                0.0           0.002631   \n",
       "2011-01-13 00:00:00+00:00                0.0          -0.005084   \n",
       "\n",
       "                           Equity(488 [ZBH])  Equity(489 [ZION])  \\\n",
       "2011-01-07 00:00:00+00:00          -0.000181           -0.010458   \n",
       "2011-01-10 00:00:00+00:00           0.007784           -0.017945   \n",
       "2011-01-11 00:00:00+00:00           0.035676            0.007467   \n",
       "2011-01-12 00:00:00+00:00           0.014741           -0.011903   \n",
       "2011-01-13 00:00:00+00:00          -0.004665           -0.009178   \n",
       "\n",
       "                           Equity(490 [ZTS])  \n",
       "2011-01-07 00:00:00+00:00                0.0  \n",
       "2011-01-10 00:00:00+00:00                0.0  \n",
       "2011-01-11 00:00:00+00:00                0.0  \n",
       "2011-01-12 00:00:00+00:00                0.0  \n",
       "2011-01-13 00:00:00+00:00                0.0  \n",
       "\n",
       "[5 rows x 490 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import utils\n",
    "\n",
    "# Get the returns for the fiven start and end date. Both dates must be valid trading dates\n",
    "returns = utils.get_returns(start_date='2011-01-05', end_date='2016-01-05')\n",
    "\n",
    "# Display the first rows of the returns\n",
    "returns.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Visualizing the Data\n",
    "\n",
    "As we san see above, the `returns` dataframe, contains the stock returns for 490 assets. Eventhough we can't make 490-dimensional plots, we can plot the data for two assets at a time. This plot willl then show us visually how correlated the stock returns are for a pair of stocks.\n",
    "\n",
    "In the code below, we use the `.plot.scatter(x, y)` method to make a scatter plot of the returns of column `x` and column `y`. The `x` and `y` parameters are both integers and idicate the number of the columns we want to plot. For example, if we want to see how correlated the stock of `AAL` and `AAPL` are, we can choose `x=1` and `y=3`, since we can see from the dataframe above that stock `AAL` corresponds to column number `1`, and stock `AAPL` corresponds to column number `3`. You are encouraged to plot different pairs of stocks to see how correlated they are.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fd9109ae588>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Set the default figure size\n",
    "plt.rcParams['figure.figsize'] = [10.0, 6.0]\n",
    "\n",
    "# Make scatter plot\n",
    "ax = returns.plot.scatter(x = 1, y = 3, grid = True, color = 'white', alpha = 0.5, linewidth = 0)\n",
    "ax.set_facecolor('lightslategray')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Correlation of Returns\n",
    "\n",
    "Apart from visualizing the correlation between stocks as we did above, we can also create a correlation dataframe that gives the correlation between every stock. In the code below, we can accomplish this using the `.corr()` method to calculate the correlation between all the paris of stocks in our `returns` dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "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>Equity(0 [A])</th>\n",
       "      <th>Equity(1 [AAL])</th>\n",
       "      <th>Equity(2 [AAP])</th>\n",
       "      <th>Equity(3 [AAPL])</th>\n",
       "      <th>Equity(4 [ABBV])</th>\n",
       "      <th>Equity(5 [ABC])</th>\n",
       "      <th>Equity(6 [ABT])</th>\n",
       "      <th>Equity(7 [ACN])</th>\n",
       "      <th>Equity(8 [ADBE])</th>\n",
       "      <th>Equity(9 [ADI])</th>\n",
       "      <th>...</th>\n",
       "      <th>Equity(481 [XL])</th>\n",
       "      <th>Equity(482 [XLNX])</th>\n",
       "      <th>Equity(483 [XOM])</th>\n",
       "      <th>Equity(484 [XRAY])</th>\n",
       "      <th>Equity(485 [XRX])</th>\n",
       "      <th>Equity(486 [XYL])</th>\n",
       "      <th>Equity(487 [YUM])</th>\n",
       "      <th>Equity(488 [ZBH])</th>\n",
       "      <th>Equity(489 [ZION])</th>\n",
       "      <th>Equity(490 [ZTS])</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Equity(0 [A])</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.327356</td>\n",
       "      <td>0.350274</td>\n",
       "      <td>0.357941</td>\n",
       "      <td>0.219881</td>\n",
       "      <td>0.443607</td>\n",
       "      <td>0.498730</td>\n",
       "      <td>0.528160</td>\n",
       "      <td>0.476105</td>\n",
       "      <td>0.488426</td>\n",
       "      <td>...</td>\n",
       "      <td>0.547015</td>\n",
       "      <td>0.448453</td>\n",
       "      <td>0.520123</td>\n",
       "      <td>0.568743</td>\n",
       "      <td>0.524692</td>\n",
       "      <td>0.333886</td>\n",
       "      <td>0.405655</td>\n",
       "      <td>0.485227</td>\n",
       "      <td>0.524167</td>\n",
       "      <td>0.186108</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Equity(1 [AAL])</th>\n",
       "      <td>0.327356</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.243475</td>\n",
       "      <td>0.227264</td>\n",
       "      <td>0.143983</td>\n",
       "      <td>0.296306</td>\n",
       "      <td>0.244604</td>\n",
       "      <td>0.269950</td>\n",
       "      <td>0.308133</td>\n",
       "      <td>0.285740</td>\n",
       "      <td>...</td>\n",
       "      <td>0.361259</td>\n",
       "      <td>0.241502</td>\n",
       "      <td>0.177152</td>\n",
       "      <td>0.356118</td>\n",
       "      <td>0.258117</td>\n",
       "      <td>0.173434</td>\n",
       "      <td>0.254696</td>\n",
       "      <td>0.289684</td>\n",
       "      <td>0.344630</td>\n",
       "      <td>0.138426</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Equity(2 [AAP])</th>\n",
       "      <td>0.350274</td>\n",
       "      <td>0.243475</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.210157</td>\n",
       "      <td>0.211495</td>\n",
       "      <td>0.290848</td>\n",
       "      <td>0.329866</td>\n",
       "      <td>0.293221</td>\n",
       "      <td>0.301972</td>\n",
       "      <td>0.309280</td>\n",
       "      <td>...</td>\n",
       "      <td>0.312257</td>\n",
       "      <td>0.289204</td>\n",
       "      <td>0.282981</td>\n",
       "      <td>0.346646</td>\n",
       "      <td>0.275858</td>\n",
       "      <td>0.268899</td>\n",
       "      <td>0.329385</td>\n",
       "      <td>0.302148</td>\n",
       "      <td>0.329689</td>\n",
       "      <td>0.150945</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Equity(3 [AAPL])</th>\n",
       "      <td>0.357941</td>\n",
       "      <td>0.227264</td>\n",
       "      <td>0.210157</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.124744</td>\n",
       "      <td>0.261941</td>\n",
       "      <td>0.260044</td>\n",
       "      <td>0.345766</td>\n",
       "      <td>0.317921</td>\n",
       "      <td>0.384693</td>\n",
       "      <td>...</td>\n",
       "      <td>0.292361</td>\n",
       "      <td>0.284832</td>\n",
       "      <td>0.320862</td>\n",
       "      <td>0.342525</td>\n",
       "      <td>0.322910</td>\n",
       "      <td>0.251614</td>\n",
       "      <td>0.297765</td>\n",
       "      <td>0.333980</td>\n",
       "      <td>0.349393</td>\n",
       "      <td>0.155476</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Equity(4 [ABBV])</th>\n",
       "      <td>0.219881</td>\n",
       "      <td>0.143983</td>\n",
       "      <td>0.211495</td>\n",
       "      <td>0.124744</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.265333</td>\n",
       "      <td>0.355565</td>\n",
       "      <td>0.214197</td>\n",
       "      <td>0.223873</td>\n",
       "      <td>0.256881</td>\n",
       "      <td>...</td>\n",
       "      <td>0.163470</td>\n",
       "      <td>0.235468</td>\n",
       "      <td>0.229631</td>\n",
       "      <td>0.247383</td>\n",
       "      <td>0.226287</td>\n",
       "      <td>0.253060</td>\n",
       "      <td>0.203928</td>\n",
       "      <td>0.304504</td>\n",
       "      <td>0.183191</td>\n",
       "      <td>0.340621</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 490 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                  Equity(0 [A])  Equity(1 [AAL])  Equity(2 [AAP])  \\\n",
       "Equity(0 [A])          1.000000         0.327356         0.350274   \n",
       "Equity(1 [AAL])        0.327356         1.000000         0.243475   \n",
       "Equity(2 [AAP])        0.350274         0.243475         1.000000   \n",
       "Equity(3 [AAPL])       0.357941         0.227264         0.210157   \n",
       "Equity(4 [ABBV])       0.219881         0.143983         0.211495   \n",
       "\n",
       "                  Equity(3 [AAPL])  Equity(4 [ABBV])  Equity(5 [ABC])  \\\n",
       "Equity(0 [A])             0.357941          0.219881         0.443607   \n",
       "Equity(1 [AAL])           0.227264          0.143983         0.296306   \n",
       "Equity(2 [AAP])           0.210157          0.211495         0.290848   \n",
       "Equity(3 [AAPL])          1.000000          0.124744         0.261941   \n",
       "Equity(4 [ABBV])          0.124744          1.000000         0.265333   \n",
       "\n",
       "                  Equity(6 [ABT])  Equity(7 [ACN])  Equity(8 [ADBE])  \\\n",
       "Equity(0 [A])            0.498730         0.528160          0.476105   \n",
       "Equity(1 [AAL])          0.244604         0.269950          0.308133   \n",
       "Equity(2 [AAP])          0.329866         0.293221          0.301972   \n",
       "Equity(3 [AAPL])         0.260044         0.345766          0.317921   \n",
       "Equity(4 [ABBV])         0.355565         0.214197          0.223873   \n",
       "\n",
       "                  Equity(9 [ADI])        ...          Equity(481 [XL])  \\\n",
       "Equity(0 [A])            0.488426        ...                  0.547015   \n",
       "Equity(1 [AAL])          0.285740        ...                  0.361259   \n",
       "Equity(2 [AAP])          0.309280        ...                  0.312257   \n",
       "Equity(3 [AAPL])         0.384693        ...                  0.292361   \n",
       "Equity(4 [ABBV])         0.256881        ...                  0.163470   \n",
       "\n",
       "                  Equity(482 [XLNX])  Equity(483 [XOM])  Equity(484 [XRAY])  \\\n",
       "Equity(0 [A])               0.448453           0.520123            0.568743   \n",
       "Equity(1 [AAL])             0.241502           0.177152            0.356118   \n",
       "Equity(2 [AAP])             0.289204           0.282981            0.346646   \n",
       "Equity(3 [AAPL])            0.284832           0.320862            0.342525   \n",
       "Equity(4 [ABBV])            0.235468           0.229631            0.247383   \n",
       "\n",
       "                  Equity(485 [XRX])  Equity(486 [XYL])  Equity(487 [YUM])  \\\n",
       "Equity(0 [A])              0.524692           0.333886           0.405655   \n",
       "Equity(1 [AAL])            0.258117           0.173434           0.254696   \n",
       "Equity(2 [AAP])            0.275858           0.268899           0.329385   \n",
       "Equity(3 [AAPL])           0.322910           0.251614           0.297765   \n",
       "Equity(4 [ABBV])           0.226287           0.253060           0.203928   \n",
       "\n",
       "                  Equity(488 [ZBH])  Equity(489 [ZION])  Equity(490 [ZTS])  \n",
       "Equity(0 [A])              0.485227            0.524167           0.186108  \n",
       "Equity(1 [AAL])            0.289684            0.344630           0.138426  \n",
       "Equity(2 [AAP])            0.302148            0.329689           0.150945  \n",
       "Equity(3 [AAPL])           0.333980            0.349393           0.155476  \n",
       "Equity(4 [ABBV])           0.304504            0.183191           0.340621  \n",
       "\n",
       "[5 rows x 490 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Display the correlation between all stock pairs in the returns dataframe\n",
    "returns.corr(method = 'pearson').head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As we can see, this is a better way to see how correlated the stock returns are than through visulaization. By looking at the table we can easily spot which pairs of stock have the highest correlation. \n",
    "\n",
    "# TODO:\n",
    "\n",
    "In the code below, make a scatter of equity `A` and equity `XOM`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fd90e1a0400>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Set the default figure size\n",
    "plt.rcParams['figure.figsize'] = [10.0, 6.0]\n",
    "\n",
    "# Make scatter plot\n",
    "ax = returns.plot.scatter(x = 0, y = 482, grid = True, color = 'white', alpha = 0.5, linewidth = 0)\n",
    "ax.set_facecolor('lightslategray')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# TODO:\n",
    "\n",
    "In the code below, write a function `get_num_components(df, var_ret)` that takes a dataframe, `df`, and a value for the desired amount of variance you want to retain from the `df` dataframe,`var_ret`. In this case, the parameter `df` should be the `returns` dataframe obtained above. The parameter  `var_ret` must be anumber between 0 and 1. The function should return the number of principal components you need to retain that amount of variance. To do this, use Scikit-Learn's PCA() class and its `.explained_variance_ratio_`. The function should also print the total amount of variance retained. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total Variance Retained:  1.0\n",
      "\n",
      "Number of Principal Components Needed:  179\n"
     ]
    }
   ],
   "source": [
    "# import resources\n",
    "import numpy as np\n",
    "from sklearn.decomposition import PCA\n",
    "# import utils\n",
    "\n",
    "def get_num_components(df, var_ret):\n",
    "    # Implement Function\n",
    "    # df_norm = utils.mean_normalize_data(df)\n",
    "    pca = PCA(n_components = df.shape[1])\n",
    "    pca.fit(df)\n",
    "    # print(pca.explained_variance_ratio_)\n",
    "    \n",
    "    needed_components = 0\n",
    "    var_sum = 0\n",
    "    \n",
    "    for i in range(0, df.shape[1]):\n",
    "        if var_sum > var_ret:\n",
    "            print('Total Variance Retained: ', pca.explained_variance_ratio_[0:].sum() )\n",
    "            return needed_components\n",
    "        else:\n",
    "            needed_components += 1\n",
    "            var_sum += pca.explained_variance_ratio_[i]        \n",
    "            \n",
    "num_components = get_num_components(returns, 0.9)\n",
    "\n",
    "print('\\nNumber of Principal Components Needed: ', num_components)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# TODO:\n",
    "\n",
    "In the previous section you calculated the number of principal compenents needed to retain a given amount of variance. As you might notice you can greatly reduce the dimensions of the data even if you retain a high level of variance (`var_ret` > 0.9). In the code below, use the number of components needed calculated in the last section, `num_components` to calculate by the percentage of dimensionality reduction. For example, if the original data was 100-dimensional, and the amount of components needed to retian a certain amount of variance is 70, then we are able to reduce the data by 30%. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "We were able to reduce the dimenionality of the data by: 63.469387755102034 percent\n"
     ]
    }
   ],
   "source": [
    "# Calculate the percentage of dimensionality reduction\n",
    "red_per = (returns.shape[1] - num_components) / returns.shape[1] * 100\n",
    "print('We were able to reduce the dimenionality of the data by:', red_per, 'percent')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Solution\n",
    "\n",
    "[Solution notebook](pca_basics_solution.ipynb)"
   ]
  }
 ],
 "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}