0501 - Outliers and Trends.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    }
   ],
   "source": [
    "%pylab inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df = pd.read_csv('data/eu_trade_sums.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>geo</th>\n",
       "      <th>trade_type</th>\n",
       "      <th>partner</th>\n",
       "      <th>2015</th>\n",
       "      <th>2014</th>\n",
       "      <th>2013</th>\n",
       "      <th>2012</th>\n",
       "      <th>2011</th>\n",
       "      <th>2010</th>\n",
       "      <th>2009</th>\n",
       "      <th>2008</th>\n",
       "      <th>2007</th>\n",
       "      <th>2006</th>\n",
       "      <th>2005</th>\n",
       "      <th>2004</th>\n",
       "      <th>2003</th>\n",
       "      <th>2002</th>\n",
       "      <th>2001</th>\n",
       "      <th>2000</th>\n",
       "      <th>1999</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AT</td>\n",
       "      <td>Export</td>\n",
       "      <td>EU28</td>\n",
       "      <td>193043.0</td>\n",
       "      <td>187548.0</td>\n",
       "      <td>184865.0</td>\n",
       "      <td>181689.0</td>\n",
       "      <td>181889.0</td>\n",
       "      <td>166299.0</td>\n",
       "      <td>143052.0</td>\n",
       "      <td>181180.0</td>\n",
       "      <td>176327.0</td>\n",
       "      <td>159517.0</td>\n",
       "      <td>147064.0</td>\n",
       "      <td>142481.0</td>\n",
       "      <td>131439.0</td>\n",
       "      <td>126666.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AT</td>\n",
       "      <td>Export</td>\n",
       "      <td>EXT_EU28</td>\n",
       "      <td>82468.0</td>\n",
       "      <td>80797.0</td>\n",
       "      <td>78903.0</td>\n",
       "      <td>77668.0</td>\n",
       "      <td>73038.0</td>\n",
       "      <td>63858.0</td>\n",
       "      <td>53376.0</td>\n",
       "      <td>65339.0</td>\n",
       "      <td>62447.0</td>\n",
       "      <td>58309.0</td>\n",
       "      <td>53871.0</td>\n",
       "      <td>46923.0</td>\n",
       "      <td>40317.0</td>\n",
       "      <td>39731.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>AT</td>\n",
       "      <td>Import</td>\n",
       "      <td>EU28</td>\n",
       "      <td>215023.0</td>\n",
       "      <td>210492.0</td>\n",
       "      <td>211444.0</td>\n",
       "      <td>212546.0</td>\n",
       "      <td>212734.0</td>\n",
       "      <td>187011.0</td>\n",
       "      <td>160875.0</td>\n",
       "      <td>197137.0</td>\n",
       "      <td>190105.0</td>\n",
       "      <td>175463.0</td>\n",
       "      <td>165908.0</td>\n",
       "      <td>160883.0</td>\n",
       "      <td>144982.0</td>\n",
       "      <td>134624.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>AT</td>\n",
       "      <td>Import</td>\n",
       "      <td>EXT_EU28</td>\n",
       "      <td>65242.0</td>\n",
       "      <td>63511.0</td>\n",
       "      <td>64554.0</td>\n",
       "      <td>65337.0</td>\n",
       "      <td>62293.0</td>\n",
       "      <td>52877.0</td>\n",
       "      <td>44262.0</td>\n",
       "      <td>53467.0</td>\n",
       "      <td>47817.0</td>\n",
       "      <td>43097.0</td>\n",
       "      <td>38657.0</td>\n",
       "      <td>31628.0</td>\n",
       "      <td>30994.0</td>\n",
       "      <td>30983.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>BE</td>\n",
       "      <td>Export</td>\n",
       "      <td>EU28</td>\n",
       "      <td>517836.0</td>\n",
       "      <td>502888.0</td>\n",
       "      <td>495145.0</td>\n",
       "      <td>486102.0</td>\n",
       "      <td>492083.0</td>\n",
       "      <td>449191.0</td>\n",
       "      <td>402557.0</td>\n",
       "      <td>493688.0</td>\n",
       "      <td>478972.0</td>\n",
       "      <td>446775.0</td>\n",
       "      <td>412912.0</td>\n",
       "      <td>380546.0</td>\n",
       "      <td>349210.0</td>\n",
       "      <td>345248.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  geo trade_type   partner      2015      2014      2013      2012      2011  \\\n",
       "0  AT     Export      EU28  193043.0  187548.0  184865.0  181689.0  181889.0   \n",
       "1  AT     Export  EXT_EU28   82468.0   80797.0   78903.0   77668.0   73038.0   \n",
       "2  AT     Import      EU28  215023.0  210492.0  211444.0  212546.0  212734.0   \n",
       "3  AT     Import  EXT_EU28   65242.0   63511.0   64554.0   65337.0   62293.0   \n",
       "4  BE     Export      EU28  517836.0  502888.0  495145.0  486102.0  492083.0   \n",
       "\n",
       "       2010      2009      2008      2007      2006      2005      2004  \\\n",
       "0  166299.0  143052.0  181180.0  176327.0  159517.0  147064.0  142481.0   \n",
       "1   63858.0   53376.0   65339.0   62447.0   58309.0   53871.0   46923.0   \n",
       "2  187011.0  160875.0  197137.0  190105.0  175463.0  165908.0  160883.0   \n",
       "3   52877.0   44262.0   53467.0   47817.0   43097.0   38657.0   31628.0   \n",
       "4  449191.0  402557.0  493688.0  478972.0  446775.0  412912.0  380546.0   \n",
       "\n",
       "       2003      2002  2001  2000  1999  \n",
       "0  131439.0  126666.0   NaN   NaN   NaN  \n",
       "1   40317.0   39731.0   NaN   NaN   NaN  \n",
       "2  144982.0  134624.0   NaN   NaN   NaN  \n",
       "3   30994.0   30983.0   NaN   NaN   NaN  \n",
       "4  349210.0  345248.0   NaN   NaN   NaN  "
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "geo            object\n",
       "trade_type     object\n",
       "partner        object\n",
       "2015          float64\n",
       "2014          float64\n",
       "2013          float64\n",
       "2012          float64\n",
       "2011          float64\n",
       "2010          float64\n",
       "2009          float64\n",
       "2008          float64\n",
       "2007          float64\n",
       "2006          float64\n",
       "2005          float64\n",
       "2004          float64\n",
       "2003          float64\n",
       "2002          float64\n",
       "2001          float64\n",
       "2000          float64\n",
       "1999          float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2015    911129.253455\n",
       "2014    872860.317369\n",
       "2013    854588.449067\n",
       "2012    858265.953440\n",
       "2011    842434.639082\n",
       "2010    756468.435997\n",
       "2009    644715.971496\n",
       "2008    798970.905808\n",
       "2007    774308.404505\n",
       "2006    724946.672056\n",
       "2005    643149.743817\n",
       "2004    594179.481775\n",
       "2003    545375.815568\n",
       "2002    541897.557364\n",
       "2001              NaN\n",
       "2000              NaN\n",
       "1999              NaN\n",
       "dtype: float64"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.std()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "yrs = [str(yr) for yr in range(2002,2016)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "df = df.set_index('geo')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f0cf6842f10>"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY0AAAEACAYAAABPiSrXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nNEREJDUlDRERSe3/A03hCz2FTHkfAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f0cf6862f90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df[(df['trade_type'] == 'Export') & \n",
    "        (df['partner'] == 'EU28')].loc[['EU28', 'UK']][yrs].T.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f0cf665f110>"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY0AAAEACAYAAABPiSrXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQ0REElNoiIhIYgoNERFJ7P8DXyfhJYNncNoAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f0cf65ce390>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df[(df['trade_type'] == 'Import') & \n",
    "        (df['partner'] == 'EU28')].loc[['EU28', 'UK']][yrs].T.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f0cf66495d0>"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY0AAAEACAYAAABPiSrXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f0cf651f550>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df[(df['trade_type'] == 'Import') & \n",
    "        (df['partner'] == 'EXT_EU28')].loc[['EU28', 'UK']][yrs].T.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f0cf63cf310>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY0AAAEACAYAAABPiSrXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nE5oFDWOMMaFZ0DDGGBOaBQ1jjDGh/X8xP3gqMEYmaAAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f0cf6343490>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df[(df['trade_type'] == 'Export') & \n",
    "        (df['partner'] == 'EXT_EU28')].loc[['EU28', 'UK']][yrs].T.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "export_df = df[(df['trade_type'] == 'Export') & (df['partner'] == 'EXT_EU28')].loc[['EU28', 'UK']][yrs].T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "export_df = export_df.rename(columns={'EU28':'EU28_TO_EXT', 'UK': 'UK_TO_EXT'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "export_df = pd.concat([export_df, df[(df['trade_type'] == 'Export') & \n",
    "        (df['partner'] == 'EU28')].loc[['EU28', 'UK']][yrs].T], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "export_df = export_df.rename(columns={'EU28':'EU28_TO_INT', 'UK': 'UK_TO_INT'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f0cf6a959d0>"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY0AAAEACAYAAABPiSrXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ns6BhjDEmaBY0jDHGBO3/AxLL2pH2y/pkAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f0cfb2b6090>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "export_df.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f0cf5f95fd0>"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYcAAAEACAYAAABYq7oeAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\najQ4iIhIjQYHERGp+QPny7wPQegtjwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f0cf5fbcc10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "export_df[['UK_TO_EXT', 'UK_TO_INT']].plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>trade_type</th>\n",
       "      <th>partner</th>\n",
       "      <th>2015</th>\n",
       "      <th>2014</th>\n",
       "      <th>2013</th>\n",
       "      <th>2012</th>\n",
       "      <th>2011</th>\n",
       "      <th>2010</th>\n",
       "      <th>2009</th>\n",
       "      <th>2008</th>\n",
       "      <th>2007</th>\n",
       "      <th>2006</th>\n",
       "      <th>2005</th>\n",
       "      <th>2004</th>\n",
       "      <th>2003</th>\n",
       "      <th>2002</th>\n",
       "      <th>2001</th>\n",
       "      <th>2000</th>\n",
       "      <th>1999</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>geo</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>AT</th>\n",
       "      <td>Export</td>\n",
       "      <td>EU28</td>\n",
       "      <td>193043.0</td>\n",
       "      <td>187548.0</td>\n",
       "      <td>184865.0</td>\n",
       "      <td>181689.0</td>\n",
       "      <td>181889.0</td>\n",
       "      <td>166299.0</td>\n",
       "      <td>143052.0</td>\n",
       "      <td>181180.0</td>\n",
       "      <td>176327.0</td>\n",
       "      <td>159517.0</td>\n",
       "      <td>147064.0</td>\n",
       "      <td>142481.0</td>\n",
       "      <td>131439.0</td>\n",
       "      <td>126666.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AT</th>\n",
       "      <td>Export</td>\n",
       "      <td>EXT_EU28</td>\n",
       "      <td>82468.0</td>\n",
       "      <td>80797.0</td>\n",
       "      <td>78903.0</td>\n",
       "      <td>77668.0</td>\n",
       "      <td>73038.0</td>\n",
       "      <td>63858.0</td>\n",
       "      <td>53376.0</td>\n",
       "      <td>65339.0</td>\n",
       "      <td>62447.0</td>\n",
       "      <td>58309.0</td>\n",
       "      <td>53871.0</td>\n",
       "      <td>46923.0</td>\n",
       "      <td>40317.0</td>\n",
       "      <td>39731.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AT</th>\n",
       "      <td>Import</td>\n",
       "      <td>EU28</td>\n",
       "      <td>215023.0</td>\n",
       "      <td>210492.0</td>\n",
       "      <td>211444.0</td>\n",
       "      <td>212546.0</td>\n",
       "      <td>212734.0</td>\n",
       "      <td>187011.0</td>\n",
       "      <td>160875.0</td>\n",
       "      <td>197137.0</td>\n",
       "      <td>190105.0</td>\n",
       "      <td>175463.0</td>\n",
       "      <td>165908.0</td>\n",
       "      <td>160883.0</td>\n",
       "      <td>144982.0</td>\n",
       "      <td>134624.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AT</th>\n",
       "      <td>Import</td>\n",
       "      <td>EXT_EU28</td>\n",
       "      <td>65242.0</td>\n",
       "      <td>63511.0</td>\n",
       "      <td>64554.0</td>\n",
       "      <td>65337.0</td>\n",
       "      <td>62293.0</td>\n",
       "      <td>52877.0</td>\n",
       "      <td>44262.0</td>\n",
       "      <td>53467.0</td>\n",
       "      <td>47817.0</td>\n",
       "      <td>43097.0</td>\n",
       "      <td>38657.0</td>\n",
       "      <td>31628.0</td>\n",
       "      <td>30994.0</td>\n",
       "      <td>30983.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BE</th>\n",
       "      <td>Export</td>\n",
       "      <td>EU28</td>\n",
       "      <td>517836.0</td>\n",
       "      <td>502888.0</td>\n",
       "      <td>495145.0</td>\n",
       "      <td>486102.0</td>\n",
       "      <td>492083.0</td>\n",
       "      <td>449191.0</td>\n",
       "      <td>402557.0</td>\n",
       "      <td>493688.0</td>\n",
       "      <td>478972.0</td>\n",
       "      <td>446775.0</td>\n",
       "      <td>412912.0</td>\n",
       "      <td>380546.0</td>\n",
       "      <td>349210.0</td>\n",
       "      <td>345248.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    trade_type   partner      2015      2014      2013      2012      2011  \\\n",
       "geo                                                                          \n",
       "AT      Export      EU28  193043.0  187548.0  184865.0  181689.0  181889.0   \n",
       "AT      Export  EXT_EU28   82468.0   80797.0   78903.0   77668.0   73038.0   \n",
       "AT      Import      EU28  215023.0  210492.0  211444.0  212546.0  212734.0   \n",
       "AT      Import  EXT_EU28   65242.0   63511.0   64554.0   65337.0   62293.0   \n",
       "BE      Export      EU28  517836.0  502888.0  495145.0  486102.0  492083.0   \n",
       "\n",
       "         2010      2009      2008      2007      2006      2005      2004  \\\n",
       "geo                                                                         \n",
       "AT   166299.0  143052.0  181180.0  176327.0  159517.0  147064.0  142481.0   \n",
       "AT    63858.0   53376.0   65339.0   62447.0   58309.0   53871.0   46923.0   \n",
       "AT   187011.0  160875.0  197137.0  190105.0  175463.0  165908.0  160883.0   \n",
       "AT    52877.0   44262.0   53467.0   47817.0   43097.0   38657.0   31628.0   \n",
       "BE   449191.0  402557.0  493688.0  478972.0  446775.0  412912.0  380546.0   \n",
       "\n",
       "         2003      2002  2001  2000  1999  \n",
       "geo                                        \n",
       "AT   131439.0  126666.0   NaN   NaN   NaN  \n",
       "AT    40317.0   39731.0   NaN   NaN   NaN  \n",
       "AT   144982.0  134624.0   NaN   NaN   NaN  \n",
       "AT    30994.0   30983.0   NaN   NaN   NaN  \n",
       "BE   349210.0  345248.0   NaN   NaN   NaN  "
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "pct_change_df = df.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "for yr in yrs:\n",
    "    pct_change_df[yr] = (df[yr] - df[str(int(yr)-1)]) / df[str(int(yr)-1)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "df = df[~df.index.isin(['EU28'])]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>trade_type</th>\n",
       "      <th>partner</th>\n",
       "      <th>2015</th>\n",
       "      <th>2014</th>\n",
       "      <th>2013</th>\n",
       "      <th>2012</th>\n",
       "      <th>2011</th>\n",
       "      <th>2010</th>\n",
       "      <th>2009</th>\n",
       "      <th>2008</th>\n",
       "      <th>2007</th>\n",
       "      <th>2006</th>\n",
       "      <th>2005</th>\n",
       "      <th>2004</th>\n",
       "      <th>2003</th>\n",
       "      <th>2002</th>\n",
       "      <th>2001</th>\n",
       "      <th>2000</th>\n",
       "      <th>1999</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>geo</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>AT</th>\n",
       "      <td>Export</td>\n",
       "      <td>EU28</td>\n",
       "      <td>0.029299</td>\n",
       "      <td>0.014513</td>\n",
       "      <td>0.017480</td>\n",
       "      <td>-0.001100</td>\n",
       "      <td>0.093747</td>\n",
       "      <td>0.162507</td>\n",
       "      <td>-0.210443</td>\n",
       "      <td>0.027523</td>\n",
       "      <td>0.105381</td>\n",
       "      <td>0.084677</td>\n",
       "      <td>0.032166</td>\n",
       "      <td>0.084009</td>\n",
       "      <td>0.037682</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AT</th>\n",
       "      <td>Export</td>\n",
       "      <td>EXT_EU28</td>\n",
       "      <td>0.020681</td>\n",
       "      <td>0.024004</td>\n",
       "      <td>0.015901</td>\n",
       "      <td>0.063392</td>\n",
       "      <td>0.143756</td>\n",
       "      <td>0.196380</td>\n",
       "      <td>-0.183091</td>\n",
       "      <td>0.046311</td>\n",
       "      <td>0.070967</td>\n",
       "      <td>0.082382</td>\n",
       "      <td>0.148072</td>\n",
       "      <td>0.163851</td>\n",
       "      <td>0.014749</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AT</th>\n",
       "      <td>Import</td>\n",
       "      <td>EU28</td>\n",
       "      <td>0.021526</td>\n",
       "      <td>-0.004502</td>\n",
       "      <td>-0.005185</td>\n",
       "      <td>-0.000884</td>\n",
       "      <td>0.137548</td>\n",
       "      <td>0.162462</td>\n",
       "      <td>-0.183943</td>\n",
       "      <td>0.036990</td>\n",
       "      <td>0.083448</td>\n",
       "      <td>0.057592</td>\n",
       "      <td>0.031234</td>\n",
       "      <td>0.109676</td>\n",
       "      <td>0.076940</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AT</th>\n",
       "      <td>Import</td>\n",
       "      <td>EXT_EU28</td>\n",
       "      <td>0.027255</td>\n",
       "      <td>-0.016157</td>\n",
       "      <td>-0.011984</td>\n",
       "      <td>0.048866</td>\n",
       "      <td>0.178074</td>\n",
       "      <td>0.194636</td>\n",
       "      <td>-0.172162</td>\n",
       "      <td>0.118159</td>\n",
       "      <td>0.109520</td>\n",
       "      <td>0.114856</td>\n",
       "      <td>0.222240</td>\n",
       "      <td>0.020456</td>\n",
       "      <td>0.000355</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>BE</th>\n",
       "      <td>Export</td>\n",
       "      <td>EU28</td>\n",
       "      <td>0.029724</td>\n",
       "      <td>0.015638</td>\n",
       "      <td>0.018603</td>\n",
       "      <td>-0.012154</td>\n",
       "      <td>0.095487</td>\n",
       "      <td>0.115844</td>\n",
       "      <td>-0.184592</td>\n",
       "      <td>0.030724</td>\n",
       "      <td>0.072065</td>\n",
       "      <td>0.082010</td>\n",
       "      <td>0.085051</td>\n",
       "      <td>0.089734</td>\n",
       "      <td>0.011476</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    trade_type   partner      2015      2014      2013      2012      2011  \\\n",
       "geo                                                                          \n",
       "AT      Export      EU28  0.029299  0.014513  0.017480 -0.001100  0.093747   \n",
       "AT      Export  EXT_EU28  0.020681  0.024004  0.015901  0.063392  0.143756   \n",
       "AT      Import      EU28  0.021526 -0.004502 -0.005185 -0.000884  0.137548   \n",
       "AT      Import  EXT_EU28  0.027255 -0.016157 -0.011984  0.048866  0.178074   \n",
       "BE      Export      EU28  0.029724  0.015638  0.018603 -0.012154  0.095487   \n",
       "\n",
       "         2010      2009      2008      2007      2006      2005      2004  \\\n",
       "geo                                                                         \n",
       "AT   0.162507 -0.210443  0.027523  0.105381  0.084677  0.032166  0.084009   \n",
       "AT   0.196380 -0.183091  0.046311  0.070967  0.082382  0.148072  0.163851   \n",
       "AT   0.162462 -0.183943  0.036990  0.083448  0.057592  0.031234  0.109676   \n",
       "AT   0.194636 -0.172162  0.118159  0.109520  0.114856  0.222240  0.020456   \n",
       "BE   0.115844 -0.184592  0.030724  0.072065  0.082010  0.085051  0.089734   \n",
       "\n",
       "         2003  2002  2001  2000  1999  \n",
       "geo                                    \n",
       "AT   0.037682   NaN   NaN   NaN   NaN  \n",
       "AT   0.014749   NaN   NaN   NaN   NaN  \n",
       "AT   0.076940   NaN   NaN   NaN   NaN  \n",
       "AT   0.000355   NaN   NaN   NaN   NaN  \n",
       "BE   0.011476   NaN   NaN   NaN   NaN  "
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pct_change_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('2002', nan)\n",
      "('2003', 0.922437290580437)\n",
      "('2004', 1.1973918362807252)\n",
      "('2005', 0.77510836760343016)\n",
      "('2006', 1.0974189015338094)\n",
      "('2007', 0.68337596565606351)\n",
      "('2008', 0.73242893213138349)\n",
      "('2009', 0.41701830776207)\n",
      "('2010', 0.77620824423939028)\n",
      "('2011', 0.64468658836173809)\n",
      "('2012', 0.40757073398065757)\n",
      "('2013', 0.50770193608178382)\n",
      "('2014', 0.69787182660594926)\n",
      "('2015', 0.65402845613708527)\n"
     ]
    }
   ],
   "source": [
    "for yr in yrs:\n",
    "    print(yr, pct_change_df[yr].max() - pct_change_df[yr].min())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.20447136781293507"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pct_change_df['2010'].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.11778424716089249"
      ]
     },
     "execution_count": 98,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pct_change_df['2010'].std()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>trade_type</th>\n",
       "      <th>partner</th>\n",
       "      <th>2015</th>\n",
       "      <th>2014</th>\n",
       "      <th>2013</th>\n",
       "      <th>2012</th>\n",
       "      <th>2011</th>\n",
       "      <th>2010</th>\n",
       "      <th>2009</th>\n",
       "      <th>2008</th>\n",
       "      <th>2007</th>\n",
       "      <th>2006</th>\n",
       "      <th>2005</th>\n",
       "      <th>2004</th>\n",
       "      <th>2003</th>\n",
       "      <th>2002</th>\n",
       "      <th>2001</th>\n",
       "      <th>2000</th>\n",
       "      <th>1999</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>geo</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>BG</th>\n",
       "      <td>Export</td>\n",
       "      <td>EXT_EU28</td>\n",
       "      <td>0.000785</td>\n",
       "      <td>-0.071965</td>\n",
       "      <td>0.045409</td>\n",
       "      <td>0.126386</td>\n",
       "      <td>0.260671</td>\n",
       "      <td>0.489159</td>\n",
       "      <td>-0.322789</td>\n",
       "      <td>0.153392</td>\n",
       "      <td>0.169818</td>\n",
       "      <td>0.266743</td>\n",
       "      <td>0.288433</td>\n",
       "      <td>0.119338</td>\n",
       "      <td>0.058218</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CZ</th>\n",
       "      <td>Import</td>\n",
       "      <td>EXT_EU28</td>\n",
       "      <td>0.102882</td>\n",
       "      <td>0.043694</td>\n",
       "      <td>-0.070514</td>\n",
       "      <td>-0.022636</td>\n",
       "      <td>0.158539</td>\n",
       "      <td>0.452684</td>\n",
       "      <td>-0.259793</td>\n",
       "      <td>0.301929</td>\n",
       "      <td>0.185348</td>\n",
       "      <td>0.266972</td>\n",
       "      <td>0.032504</td>\n",
       "      <td>-0.154460</td>\n",
       "      <td>0.107936</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LT</th>\n",
       "      <td>Export</td>\n",
       "      <td>EXT_EU28</td>\n",
       "      <td>-0.191402</td>\n",
       "      <td>0.006403</td>\n",
       "      <td>0.198597</td>\n",
       "      <td>0.172429</td>\n",
       "      <td>0.276142</td>\n",
       "      <td>0.449822</td>\n",
       "      <td>-0.340340</td>\n",
       "      <td>0.449244</td>\n",
       "      <td>0.075822</td>\n",
       "      <td>0.257226</td>\n",
       "      <td>0.331423</td>\n",
       "      <td>0.067060</td>\n",
       "      <td>0.348454</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    trade_type   partner      2015      2014      2013      2012      2011  \\\n",
       "geo                                                                          \n",
       "BG      Export  EXT_EU28  0.000785 -0.071965  0.045409  0.126386  0.260671   \n",
       "CZ      Import  EXT_EU28  0.102882  0.043694 -0.070514 -0.022636  0.158539   \n",
       "LT      Export  EXT_EU28 -0.191402  0.006403  0.198597  0.172429  0.276142   \n",
       "\n",
       "         2010      2009      2008      2007      2006      2005      2004  \\\n",
       "geo                                                                         \n",
       "BG   0.489159 -0.322789  0.153392  0.169818  0.266743  0.288433  0.119338   \n",
       "CZ   0.452684 -0.259793  0.301929  0.185348  0.266972  0.032504 -0.154460   \n",
       "LT   0.449822 -0.340340  0.449244  0.075822  0.257226  0.331423  0.067060   \n",
       "\n",
       "         2003  2002  2001  2000  1999  \n",
       "geo                                    \n",
       "BG   0.058218   NaN   NaN   NaN   NaN  \n",
       "CZ   0.107936   NaN   NaN   NaN   NaN  \n",
       "LT   0.348454   NaN   NaN   NaN   NaN  "
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pct_change_df[pct_change_df['2010'].abs() >= (pct_change_df['2010'].mean() + 2*pct_change_df['2010'].std())]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "geo\n",
       "LU   -0.287050\n",
       "LU   -0.118591\n",
       "EL   -0.113044\n",
       "HR   -0.045526\n",
       "IE   -0.042371\n",
       "IE    0.003527\n",
       "IE    0.039697\n",
       "DK    0.047805\n",
       "DK    0.050215\n",
       "DK    0.053122\n",
       "HR    0.062120\n",
       "EL    0.093680\n",
       "FR    0.106570\n",
       "PT    0.109449\n",
       "ES    0.110758\n",
       "BG    0.111965\n",
       "MT    0.112542\n",
       "BE    0.115844\n",
       "CY    0.117683\n",
       "FR    0.123650\n",
       "EL    0.132565\n",
       "FI    0.136728\n",
       "DE    0.139136\n",
       "BE    0.139652\n",
       "IE    0.142192\n",
       "DK    0.146256\n",
       "SI    0.146373\n",
       "IT    0.147951\n",
       "SI    0.148808\n",
       "LU    0.157511\n",
       "        ...   \n",
       "PL    0.266329\n",
       "NL    0.268815\n",
       "ES    0.277191\n",
       "BE    0.281675\n",
       "EL    0.284633\n",
       "LT    0.288320\n",
       "SE    0.289761\n",
       "LV    0.294582\n",
       "UK    0.294665\n",
       "MT    0.294668\n",
       "CZ    0.306317\n",
       "IT    0.312230\n",
       "EE    0.312303\n",
       "SE    0.315566\n",
       "LV    0.316998\n",
       "HU    0.318835\n",
       "SI    0.325969\n",
       "EE    0.329750\n",
       "PL    0.336837\n",
       "SE    0.344466\n",
       "SK    0.351815\n",
       "MT    0.357491\n",
       "SK    0.372937\n",
       "RO    0.387858\n",
       "EE    0.389522\n",
       "MT    0.420200\n",
       "LT    0.427507\n",
       "LT    0.449822\n",
       "CZ    0.452684\n",
       "BG    0.489159\n",
       "Name: 2010, dtype: float64"
      ]
     },
     "execution_count": 100,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pct_change_df['2010'].sort_values()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>trade_type</th>\n",
       "      <th>partner</th>\n",
       "      <th>2015</th>\n",
       "      <th>2014</th>\n",
       "      <th>2013</th>\n",
       "      <th>2012</th>\n",
       "      <th>2011</th>\n",
       "      <th>2010</th>\n",
       "      <th>2009</th>\n",
       "      <th>2008</th>\n",
       "      <th>2007</th>\n",
       "      <th>2006</th>\n",
       "      <th>2005</th>\n",
       "      <th>2004</th>\n",
       "      <th>2003</th>\n",
       "      <th>2002</th>\n",
       "      <th>2001</th>\n",
       "      <th>2000</th>\n",
       "      <th>1999</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>geo</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>EL</th>\n",
       "      <td>Import</td>\n",
       "      <td>EU28</td>\n",
       "      <td>-0.011554</td>\n",
       "      <td>0.052153</td>\n",
       "      <td>-0.024641</td>\n",
       "      <td>-0.092378</td>\n",
       "      <td>-0.064558</td>\n",
       "      <td>-0.113044</td>\n",
       "      <td>-0.167735</td>\n",
       "      <td>0.034348</td>\n",
       "      <td>0.121114</td>\n",
       "      <td>0.105525</td>\n",
       "      <td>-0.002684</td>\n",
       "      <td>0.223572</td>\n",
       "      <td>0.242409</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HR</th>\n",
       "      <td>Import</td>\n",
       "      <td>EU28</td>\n",
       "      <td>0.103244</td>\n",
       "      <td>0.178629</td>\n",
       "      <td>0.094336</td>\n",
       "      <td>0.006855</td>\n",
       "      <td>0.104890</td>\n",
       "      <td>-0.045526</td>\n",
       "      <td>-0.285067</td>\n",
       "      <td>0.094356</td>\n",
       "      <td>0.061801</td>\n",
       "      <td>0.131649</td>\n",
       "      <td>0.070830</td>\n",
       "      <td>0.033353</td>\n",
       "      <td>0.126327</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>IE</th>\n",
       "      <td>Import</td>\n",
       "      <td>EXT_EU28</td>\n",
       "      <td>0.095668</td>\n",
       "      <td>0.146735</td>\n",
       "      <td>0.097808</td>\n",
       "      <td>0.108564</td>\n",
       "      <td>-0.016647</td>\n",
       "      <td>-0.042371</td>\n",
       "      <td>-0.097875</td>\n",
       "      <td>-0.064441</td>\n",
       "      <td>0.001091</td>\n",
       "      <td>-0.000545</td>\n",
       "      <td>0.082031</td>\n",
       "      <td>-0.042978</td>\n",
       "      <td>-0.028298</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LU</th>\n",
       "      <td>Export</td>\n",
       "      <td>EU28</td>\n",
       "      <td>0.092843</td>\n",
       "      <td>0.062869</td>\n",
       "      <td>-0.027817</td>\n",
       "      <td>-0.037380</td>\n",
       "      <td>0.021101</td>\n",
       "      <td>-0.118591</td>\n",
       "      <td>-0.135602</td>\n",
       "      <td>0.051926</td>\n",
       "      <td>-0.102930</td>\n",
       "      <td>0.190753</td>\n",
       "      <td>0.166059</td>\n",
       "      <td>0.122882</td>\n",
       "      <td>0.100969</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LU</th>\n",
       "      <td>Import</td>\n",
       "      <td>EXT_EU28</td>\n",
       "      <td>0.452790</td>\n",
       "      <td>-0.048928</td>\n",
       "      <td>-0.136559</td>\n",
       "      <td>0.279079</td>\n",
       "      <td>0.032977</td>\n",
       "      <td>-0.287050</td>\n",
       "      <td>-0.062342</td>\n",
       "      <td>0.028338</td>\n",
       "      <td>-0.153985</td>\n",
       "      <td>0.269407</td>\n",
       "      <td>0.294330</td>\n",
       "      <td>0.201769</td>\n",
       "      <td>0.441710</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    trade_type   partner      2015      2014      2013      2012      2011  \\\n",
       "geo                                                                          \n",
       "EL      Import      EU28 -0.011554  0.052153 -0.024641 -0.092378 -0.064558   \n",
       "HR      Import      EU28  0.103244  0.178629  0.094336  0.006855  0.104890   \n",
       "IE      Import  EXT_EU28  0.095668  0.146735  0.097808  0.108564 -0.016647   \n",
       "LU      Export      EU28  0.092843  0.062869 -0.027817 -0.037380  0.021101   \n",
       "LU      Import  EXT_EU28  0.452790 -0.048928 -0.136559  0.279079  0.032977   \n",
       "\n",
       "         2010      2009      2008      2007      2006      2005      2004  \\\n",
       "geo                                                                         \n",
       "EL  -0.113044 -0.167735  0.034348  0.121114  0.105525 -0.002684  0.223572   \n",
       "HR  -0.045526 -0.285067  0.094356  0.061801  0.131649  0.070830  0.033353   \n",
       "IE  -0.042371 -0.097875 -0.064441  0.001091 -0.000545  0.082031 -0.042978   \n",
       "LU  -0.118591 -0.135602  0.051926 -0.102930  0.190753  0.166059  0.122882   \n",
       "LU  -0.287050 -0.062342  0.028338 -0.153985  0.269407  0.294330  0.201769   \n",
       "\n",
       "         2003  2002  2001  2000  1999  \n",
       "geo                                    \n",
       "EL   0.242409   NaN   NaN   NaN   NaN  \n",
       "HR   0.126327   NaN   NaN   NaN   NaN  \n",
       "IE  -0.028298   NaN   NaN   NaN   NaN  \n",
       "LU   0.100969   NaN   NaN   NaN   NaN  \n",
       "LU   0.441710   NaN   NaN   NaN   NaN  "
      ]
     },
     "execution_count": 101,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pct_change_df[pct_change_df['2010'] < 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>trade_type</th>\n",
       "      <th>partner</th>\n",
       "      <th>2015</th>\n",
       "      <th>2014</th>\n",
       "      <th>2013</th>\n",
       "      <th>2012</th>\n",
       "      <th>2011</th>\n",
       "      <th>2010</th>\n",
       "      <th>2009</th>\n",
       "      <th>2008</th>\n",
       "      <th>2007</th>\n",
       "      <th>2006</th>\n",
       "      <th>2005</th>\n",
       "      <th>2004</th>\n",
       "      <th>2003</th>\n",
       "      <th>2002</th>\n",
       "      <th>2001</th>\n",
       "      <th>2000</th>\n",
       "      <th>1999</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>geo</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>BG</th>\n",
       "      <td>Export</td>\n",
       "      <td>EXT_EU28</td>\n",
       "      <td>0.000785</td>\n",
       "      <td>-0.071965</td>\n",
       "      <td>0.045409</td>\n",
       "      <td>0.126386</td>\n",
       "      <td>0.260671</td>\n",
       "      <td>0.489159</td>\n",
       "      <td>-0.322789</td>\n",
       "      <td>0.153392</td>\n",
       "      <td>0.169818</td>\n",
       "      <td>0.266743</td>\n",
       "      <td>0.288433</td>\n",
       "      <td>0.119338</td>\n",
       "      <td>0.058218</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CZ</th>\n",
       "      <td>Import</td>\n",
       "      <td>EXT_EU28</td>\n",
       "      <td>0.102882</td>\n",
       "      <td>0.043694</td>\n",
       "      <td>-0.070514</td>\n",
       "      <td>-0.022636</td>\n",
       "      <td>0.158539</td>\n",
       "      <td>0.452684</td>\n",
       "      <td>-0.259793</td>\n",
       "      <td>0.301929</td>\n",
       "      <td>0.185348</td>\n",
       "      <td>0.266972</td>\n",
       "      <td>0.032504</td>\n",
       "      <td>-0.154460</td>\n",
       "      <td>0.107936</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LT</th>\n",
       "      <td>Export</td>\n",
       "      <td>EXT_EU28</td>\n",
       "      <td>-0.191402</td>\n",
       "      <td>0.006403</td>\n",
       "      <td>0.198597</td>\n",
       "      <td>0.172429</td>\n",
       "      <td>0.276142</td>\n",
       "      <td>0.449822</td>\n",
       "      <td>-0.340340</td>\n",
       "      <td>0.449244</td>\n",
       "      <td>0.075822</td>\n",
       "      <td>0.257226</td>\n",
       "      <td>0.331423</td>\n",
       "      <td>0.067060</td>\n",
       "      <td>0.348454</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LT</th>\n",
       "      <td>Import</td>\n",
       "      <td>EXT_EU28</td>\n",
       "      <td>-0.058658</td>\n",
       "      <td>-0.143427</td>\n",
       "      <td>-0.013144</td>\n",
       "      <td>0.067575</td>\n",
       "      <td>0.288577</td>\n",
       "      <td>0.427507</td>\n",
       "      <td>-0.401784</td>\n",
       "      <td>0.592402</td>\n",
       "      <td>-0.019068</td>\n",
       "      <td>0.135555</td>\n",
       "      <td>0.393113</td>\n",
       "      <td>-0.029152</td>\n",
       "      <td>0.089929</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MT</th>\n",
       "      <td>Import</td>\n",
       "      <td>EXT_EU28</td>\n",
       "      <td>-0.054859</td>\n",
       "      <td>0.427825</td>\n",
       "      <td>0.138429</td>\n",
       "      <td>-0.004228</td>\n",
       "      <td>0.038191</td>\n",
       "      <td>0.420200</td>\n",
       "      <td>-0.055915</td>\n",
       "      <td>-0.068020</td>\n",
       "      <td>-0.092583</td>\n",
       "      <td>0.393204</td>\n",
       "      <td>-0.078005</td>\n",
       "      <td>-0.138767</td>\n",
       "      <td>0.019652</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    trade_type   partner      2015      2014      2013      2012      2011  \\\n",
       "geo                                                                          \n",
       "BG      Export  EXT_EU28  0.000785 -0.071965  0.045409  0.126386  0.260671   \n",
       "CZ      Import  EXT_EU28  0.102882  0.043694 -0.070514 -0.022636  0.158539   \n",
       "LT      Export  EXT_EU28 -0.191402  0.006403  0.198597  0.172429  0.276142   \n",
       "LT      Import  EXT_EU28 -0.058658 -0.143427 -0.013144  0.067575  0.288577   \n",
       "MT      Import  EXT_EU28 -0.054859  0.427825  0.138429 -0.004228  0.038191   \n",
       "\n",
       "         2010      2009      2008      2007      2006      2005      2004  \\\n",
       "geo                                                                         \n",
       "BG   0.489159 -0.322789  0.153392  0.169818  0.266743  0.288433  0.119338   \n",
       "CZ   0.452684 -0.259793  0.301929  0.185348  0.266972  0.032504 -0.154460   \n",
       "LT   0.449822 -0.340340  0.449244  0.075822  0.257226  0.331423  0.067060   \n",
       "LT   0.427507 -0.401784  0.592402 -0.019068  0.135555  0.393113 -0.029152   \n",
       "MT   0.420200 -0.055915 -0.068020 -0.092583  0.393204 -0.078005 -0.138767   \n",
       "\n",
       "         2003  2002  2001  2000  1999  \n",
       "geo                                    \n",
       "BG   0.058218   NaN   NaN   NaN   NaN  \n",
       "CZ   0.107936   NaN   NaN   NaN   NaN  \n",
       "LT   0.348454   NaN   NaN   NaN   NaN  \n",
       "LT   0.089929   NaN   NaN   NaN   NaN  \n",
       "MT   0.019652   NaN   NaN   NaN   NaN  "
      ]
     },
     "execution_count": 102,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pct_change_df[pct_change_df['2010'] > .4]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>trade_type</th>\n",
       "      <th>partner</th>\n",
       "      <th>2015</th>\n",
       "      <th>2014</th>\n",
       "      <th>2013</th>\n",
       "      <th>2012</th>\n",
       "      <th>2011</th>\n",
       "      <th>2010</th>\n",
       "      <th>2009</th>\n",
       "      <th>2008</th>\n",
       "      <th>2007</th>\n",
       "      <th>2006</th>\n",
       "      <th>2005</th>\n",
       "      <th>2004</th>\n",
       "      <th>2003</th>\n",
       "      <th>2002</th>\n",
       "      <th>2001</th>\n",
       "      <th>2000</th>\n",
       "      <th>1999</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>geo</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>LT</th>\n",
       "      <td>Export</td>\n",
       "      <td>EU28</td>\n",
       "      <td>0.054537</td>\n",
       "      <td>-0.018696</td>\n",
       "      <td>-0.022477</td>\n",
       "      <td>0.125611</td>\n",
       "      <td>0.294970</td>\n",
       "      <td>0.258281</td>\n",
       "      <td>-0.217510</td>\n",
       "      <td>0.196277</td>\n",
       "      <td>0.130453</td>\n",
       "      <td>0.150301</td>\n",
       "      <td>0.240060</td>\n",
       "      <td>0.300078</td>\n",
       "      <td>0.007946</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LT</th>\n",
       "      <td>Export</td>\n",
       "      <td>EXT_EU28</td>\n",
       "      <td>-0.191402</td>\n",
       "      <td>0.006403</td>\n",
       "      <td>0.198597</td>\n",
       "      <td>0.172429</td>\n",
       "      <td>0.276142</td>\n",
       "      <td>0.449822</td>\n",
       "      <td>-0.340340</td>\n",
       "      <td>0.449244</td>\n",
       "      <td>0.075822</td>\n",
       "      <td>0.257226</td>\n",
       "      <td>0.331423</td>\n",
       "      <td>0.067060</td>\n",
       "      <td>0.348454</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LT</th>\n",
       "      <td>Import</td>\n",
       "      <td>EU28</td>\n",
       "      <td>0.001737</td>\n",
       "      <td>0.074232</td>\n",
       "      <td>0.102291</td>\n",
       "      <td>0.107067</td>\n",
       "      <td>0.296293</td>\n",
       "      <td>0.288320</td>\n",
       "      <td>-0.362849</td>\n",
       "      <td>-0.000452</td>\n",
       "      <td>0.257355</td>\n",
       "      <td>0.302454</td>\n",
       "      <td>0.175332</td>\n",
       "      <td>0.322052</td>\n",
       "      <td>0.057089</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LT</th>\n",
       "      <td>Import</td>\n",
       "      <td>EXT_EU28</td>\n",
       "      <td>-0.058658</td>\n",
       "      <td>-0.143427</td>\n",
       "      <td>-0.013144</td>\n",
       "      <td>0.067575</td>\n",
       "      <td>0.288577</td>\n",
       "      <td>0.427507</td>\n",
       "      <td>-0.401784</td>\n",
       "      <td>0.592402</td>\n",
       "      <td>-0.019068</td>\n",
       "      <td>0.135555</td>\n",
       "      <td>0.393113</td>\n",
       "      <td>-0.029152</td>\n",
       "      <td>0.089929</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LU</th>\n",
       "      <td>Export</td>\n",
       "      <td>EU28</td>\n",
       "      <td>0.092843</td>\n",
       "      <td>0.062869</td>\n",
       "      <td>-0.027817</td>\n",
       "      <td>-0.037380</td>\n",
       "      <td>0.021101</td>\n",
       "      <td>-0.118591</td>\n",
       "      <td>-0.135602</td>\n",
       "      <td>0.051926</td>\n",
       "      <td>-0.102930</td>\n",
       "      <td>0.190753</td>\n",
       "      <td>0.166059</td>\n",
       "      <td>0.122882</td>\n",
       "      <td>0.100969</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LU</th>\n",
       "      <td>Export</td>\n",
       "      <td>EXT_EU28</td>\n",
       "      <td>-0.015455</td>\n",
       "      <td>-0.042497</td>\n",
       "      <td>-0.145012</td>\n",
       "      <td>0.039804</td>\n",
       "      <td>0.233666</td>\n",
       "      <td>0.239938</td>\n",
       "      <td>-0.037258</td>\n",
       "      <td>-0.012751</td>\n",
       "      <td>0.041635</td>\n",
       "      <td>0.216216</td>\n",
       "      <td>0.274851</td>\n",
       "      <td>0.011213</td>\n",
       "      <td>-0.008734</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LU</th>\n",
       "      <td>Import</td>\n",
       "      <td>EU28</td>\n",
       "      <td>-0.064594</td>\n",
       "      <td>0.002399</td>\n",
       "      <td>-0.030335</td>\n",
       "      <td>-0.021436</td>\n",
       "      <td>0.126386</td>\n",
       "      <td>0.157511</td>\n",
       "      <td>-0.205653</td>\n",
       "      <td>0.083402</td>\n",
       "      <td>-0.011737</td>\n",
       "      <td>0.158897</td>\n",
       "      <td>0.074815</td>\n",
       "      <td>0.106157</td>\n",
       "      <td>-0.006730</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>LU</th>\n",
       "      <td>Import</td>\n",
       "      <td>EXT_EU28</td>\n",
       "      <td>0.452790</td>\n",
       "      <td>-0.048928</td>\n",
       "      <td>-0.136559</td>\n",
       "      <td>0.279079</td>\n",
       "      <td>0.032977</td>\n",
       "      <td>-0.287050</td>\n",
       "      <td>-0.062342</td>\n",
       "      <td>0.028338</td>\n",
       "      <td>-0.153985</td>\n",
       "      <td>0.269407</td>\n",
       "      <td>0.294330</td>\n",
       "      <td>0.201769</td>\n",
       "      <td>0.441710</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    trade_type   partner      2015      2014      2013      2012      2011  \\\n",
       "geo                                                                          \n",
       "LT      Export      EU28  0.054537 -0.018696 -0.022477  0.125611  0.294970   \n",
       "LT      Export  EXT_EU28 -0.191402  0.006403  0.198597  0.172429  0.276142   \n",
       "LT      Import      EU28  0.001737  0.074232  0.102291  0.107067  0.296293   \n",
       "LT      Import  EXT_EU28 -0.058658 -0.143427 -0.013144  0.067575  0.288577   \n",
       "LU      Export      EU28  0.092843  0.062869 -0.027817 -0.037380  0.021101   \n",
       "LU      Export  EXT_EU28 -0.015455 -0.042497 -0.145012  0.039804  0.233666   \n",
       "LU      Import      EU28 -0.064594  0.002399 -0.030335 -0.021436  0.126386   \n",
       "LU      Import  EXT_EU28  0.452790 -0.048928 -0.136559  0.279079  0.032977   \n",
       "\n",
       "         2010      2009      2008      2007      2006      2005      2004  \\\n",
       "geo                                                                         \n",
       "LT   0.258281 -0.217510  0.196277  0.130453  0.150301  0.240060  0.300078   \n",
       "LT   0.449822 -0.340340  0.449244  0.075822  0.257226  0.331423  0.067060   \n",
       "LT   0.288320 -0.362849 -0.000452  0.257355  0.302454  0.175332  0.322052   \n",
       "LT   0.427507 -0.401784  0.592402 -0.019068  0.135555  0.393113 -0.029152   \n",
       "LU  -0.118591 -0.135602  0.051926 -0.102930  0.190753  0.166059  0.122882   \n",
       "LU   0.239938 -0.037258 -0.012751  0.041635  0.216216  0.274851  0.011213   \n",
       "LU   0.157511 -0.205653  0.083402 -0.011737  0.158897  0.074815  0.106157   \n",
       "LU  -0.287050 -0.062342  0.028338 -0.153985  0.269407  0.294330  0.201769   \n",
       "\n",
       "         2003  2002  2001  2000  1999  \n",
       "geo                                    \n",
       "LT   0.007946   NaN   NaN   NaN   NaN  \n",
       "LT   0.348454   NaN   NaN   NaN   NaN  \n",
       "LT   0.057089   NaN   NaN   NaN   NaN  \n",
       "LT   0.089929   NaN   NaN   NaN   NaN  \n",
       "LU   0.100969   NaN   NaN   NaN   NaN  \n",
       "LU  -0.008734   NaN   NaN   NaN   NaN  \n",
       "LU  -0.006730   NaN   NaN   NaN   NaN  \n",
       "LU   0.441710   NaN   NaN   NaN   NaN  "
      ]
     },
     "execution_count": 103,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pct_change_df.loc[['LT', 'LU']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "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": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}