Skip to content

Instantly share code, notes, and snippets.

@pjbull

pjbull/World Bank - Getting Started.ipynb

Last active April 18, 2017 21:47
Show Gist options
  • Save pjbull/c3b7a16fe5a690813c09 to your computer and use it in GitHub Desktop.
Save pjbull/c3b7a16fe5a690813c09 to your computer and use it in GitHub Desktop.
Download ZIP
World Bank - Getting Started
Display the source blob
Display the rendered blob
Raw
World Bank - Getting Started.ipynb
{
"metadata": {
"name": "",
"signature": "sha256:98e3bd9fc353df1c8c4ca1373ea0257f14eb8579972e5e35873695811a72baf1"
},
"nbformat": 3,
"nbformat_minor": 0,
"worksheets": [
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Getting started with the Millennium Development Goals challenge\n",
"\n",
"-----------\n",
"\n",
"### This notebook is available for [viewing](http://nbviewer.ipython.org/gist/pjbull/c3b7a16fe5a690813c09) and [download](https://gist.githubusercontent.com/pjbull/c3b7a16fe5a690813c09/raw/5a92f81bb230742e9bca853222e2653ff641a5a2/World%20Bank%20-%20Getting%20Started.ipynb) NBViewer and as a Github gist.\n",
"\n",
"----------\n",
"\n",
"People have been a little reticent to start digging in to the [Millenium Development Goals](http://www.drivendata.org/competitions/1/) challenge. We thought a little sample code might help get things started.\n",
"\n",
"The UN measures progress towards Millennium Development Goals using macroeconomic indicators such as percent of the population making over one dollar per day. Each goal has one or more targets, and each target has one or more indicators. Of the 60 indicators in all, we've chosen a subset to focus on for this challenge. Your task is to predict the change in these indicators one year and five years into the \"future\" from the year 2007.\n",
"\n",
"Predicting future progress will help us to understand how we achieve these goals by uncovering complex relations between these goals and other economic indicators. The UN set 2015 as the target for measurable progress. Given the data from 1972 - 2007, you need to predict a specific indicator for each of these goals in 2008 and 2012.\n",
"\n",
"This notebook contains three sections\n",
"\n",
" - [Digging into the Data](#digging)\n",
" - [Making a simple model](#model)\n",
" - [Starting to think about correlations](#correlations)\n",
" \n",
"We'll get started by loading the python modules in our analysis toolkit."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"%matplotlib inline\n",
"\n",
"# data manipulation and modeling\n",
"import numpy as np\n",
"import pandas as pd\n",
"import statsmodels.api as sm\n",
"\n",
"# graphix\n",
"import matplotlib.pyplot as plt\n",
"import prettyplotlib as pplt\n",
"import seaborn as sns\n",
"import statsmodels.graphics.tsaplots as tsaplots\n",
"\n",
"# utility\n",
"import os\n",
"\n",
"# notebook parameters\n",
"pd.set_option('display.max_columns', 40) # number of columns in training set\n",
"plt.rcParams['figure.figsize'] = (14.0, 8.0)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 1
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"<a id=\"digging\"></a>\n",
"\n",
"# Digging into the data\n",
"\n",
"---------------\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Time to go get that data. We've downloaded the data from [the competition data download page](http://www.drivendata.org/competitions/1/data/). We've put it in a folder called `data` in the same directory as this notebook. Now we can load in the data into the IPython notebook. There are two files here:\n",
"\n",
" - **Training Set** - This contains all of the timeseries values for 241 countries, for 1337 indicators from 1970 - 2007. We've downloaded and unzipped the archive in our `data` folder, so we are working with the CSV file.\n",
" \n",
" - **Submission Rows** - This contains the rows and format for our submission. We need to predict the values in 2008 and 2012 for the time series with IDs in this file."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"training_data = pd.read_csv(\"data/TrainingSet.csv\", index_col=0)\n",
"submission_labels = pd.read_csv(\"data/SubmissionRows.csv\", index_col=0)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 2
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"training_data.head()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>1972 [YR1972]</th>\n",
" <th>1973 [YR1973]</th>\n",
" <th>1974 [YR1974]</th>\n",
" <th>1975 [YR1975]</th>\n",
" <th>1976 [YR1976]</th>\n",
" <th>1977 [YR1977]</th>\n",
" <th>1978 [YR1978]</th>\n",
" <th>1979 [YR1979]</th>\n",
" <th>1980 [YR1980]</th>\n",
" <th>1981 [YR1981]</th>\n",
" <th>1982 [YR1982]</th>\n",
" <th>1983 [YR1983]</th>\n",
" <th>1984 [YR1984]</th>\n",
" <th>1985 [YR1985]</th>\n",
" <th>1986 [YR1986]</th>\n",
" <th>1987 [YR1987]</th>\n",
" <th>1988 [YR1988]</th>\n",
" <th>1989 [YR1989]</th>\n",
" <th>1990 [YR1990]</th>\n",
" <th>1991 [YR1991]</th>\n",
" <th>1992 [YR1992]</th>\n",
" <th>1993 [YR1993]</th>\n",
" <th>1994 [YR1994]</th>\n",
" <th>1995 [YR1995]</th>\n",
" <th>1996 [YR1996]</th>\n",
" <th>1997 [YR1997]</th>\n",
" <th>1998 [YR1998]</th>\n",
" <th>1999 [YR1999]</th>\n",
" <th>2000 [YR2000]</th>\n",
" <th>2001 [YR2001]</th>\n",
" <th>2002 [YR2002]</th>\n",
" <th>2003 [YR2003]</th>\n",
" <th>2004 [YR2004]</th>\n",
" <th>2005 [YR2005]</th>\n",
" <th>2006 [YR2006]</th>\n",
" <th>2007 [YR2007]</th>\n",
" <th>Country Name</th>\n",
" <th>Series Code</th>\n",
" <th>Series Name</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td> 3.769214</td>\n",
" <td> Afghanistan</td>\n",
" <td> allsi.bi_q1</td>\n",
" <td> (%) Benefits held by 1st 20% population - All ...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td> 7.027746</td>\n",
" <td> Afghanistan</td>\n",
" <td> allsp.bi_q1</td>\n",
" <td> (%) Benefits held by 1st 20% population - All ...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td> 8.244887</td>\n",
" <td> Afghanistan</td>\n",
" <td> allsa.bi_q1</td>\n",
" <td> (%) Benefits held by 1st 20% population - All ...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td> 12.933105</td>\n",
" <td> Afghanistan</td>\n",
" <td> allsi.gen_pop</td>\n",
" <td> (%) Generosity of All Social Insurance</td>\n",
" </tr>\n",
" <tr>\n",
" <th>5</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" <td> 18.996814</td>\n",
" <td> Afghanistan</td>\n",
" <td> allsp.gen_pop</td>\n",
" <td> (%) Generosity of All Social Protection</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>5 rows \u00d7 39 columns</p>\n",
"</div>"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 3,
"text": [
" 1972 [YR1972] 1973 [YR1973] 1974 [YR1974] 1975 [YR1975] 1976 [YR1976] \\\n",
"0 NaN NaN NaN NaN NaN \n",
"1 NaN NaN NaN NaN NaN \n",
"2 NaN NaN NaN NaN NaN \n",
"4 NaN NaN NaN NaN NaN \n",
"5 NaN NaN NaN NaN NaN \n",
"\n",
" 1977 [YR1977] 1978 [YR1978] 1979 [YR1979] 1980 [YR1980] 1981 [YR1981] \\\n",
"0 NaN NaN NaN NaN NaN \n",
"1 NaN NaN NaN NaN NaN \n",
"2 NaN NaN NaN NaN NaN \n",
"4 NaN NaN NaN NaN NaN \n",
"5 NaN NaN NaN NaN NaN \n",
"\n",
" 1982 [YR1982] 1983 [YR1983] 1984 [YR1984] 1985 [YR1985] 1986 [YR1986] \\\n",
"0 NaN NaN NaN NaN NaN \n",
"1 NaN NaN NaN NaN NaN \n",
"2 NaN NaN NaN NaN NaN \n",
"4 NaN NaN NaN NaN NaN \n",
"5 NaN NaN NaN NaN NaN \n",
"\n",
" 1987 [YR1987] 1988 [YR1988] 1989 [YR1989] 1990 [YR1990] 1991 [YR1991] \\\n",
"0 NaN NaN NaN NaN NaN \n",
"1 NaN NaN NaN NaN NaN \n",
"2 NaN NaN NaN NaN NaN \n",
"4 NaN NaN NaN NaN NaN \n",
"5 NaN NaN NaN NaN NaN \n",
"\n",
" 1992 [YR1992] 1993 [YR1993] 1994 [YR1994] 1995 [YR1995] 1996 [YR1996] \\\n",
"0 NaN NaN NaN NaN NaN \n",
"1 NaN NaN NaN NaN NaN \n",
"2 NaN NaN NaN NaN NaN \n",
"4 NaN NaN NaN NaN NaN \n",
"5 NaN NaN NaN NaN NaN \n",
"\n",
" 1997 [YR1997] 1998 [YR1998] 1999 [YR1999] 2000 [YR2000] 2001 [YR2001] \\\n",
"0 NaN NaN NaN NaN NaN \n",
"1 NaN NaN NaN NaN NaN \n",
"2 NaN NaN NaN NaN NaN \n",
"4 NaN NaN NaN NaN NaN \n",
"5 NaN NaN NaN NaN NaN \n",
"\n",
" 2002 [YR2002] 2003 [YR2003] 2004 [YR2004] 2005 [YR2005] 2006 [YR2006] \\\n",
"0 NaN NaN NaN NaN NaN \n",
"1 NaN NaN NaN NaN NaN \n",
"2 NaN NaN NaN NaN NaN \n",
"4 NaN NaN NaN NaN NaN \n",
"5 NaN NaN NaN NaN NaN \n",
"\n",
" 2007 [YR2007] Country Name Series Code \\\n",
"0 3.769214 Afghanistan allsi.bi_q1 \n",
"1 7.027746 Afghanistan allsp.bi_q1 \n",
"2 8.244887 Afghanistan allsa.bi_q1 \n",
"4 12.933105 Afghanistan allsi.gen_pop \n",
"5 18.996814 Afghanistan allsp.gen_pop \n",
"\n",
" Series Name \n",
"0 (%) Benefits held by 1st 20% population - All ... \n",
"1 (%) Benefits held by 1st 20% population - All ... \n",
"2 (%) Benefits held by 1st 20% population - All ... \n",
"4 (%) Generosity of All Social Insurance \n",
"5 (%) Generosity of All Social Protection \n",
"\n",
"[5 rows x 39 columns]"
]
}
],
"prompt_number": 3
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Poking around, we can see that there are entries for 1972 - 2007 for many different countries for many different macroeconomic indicators. Also, **it's important to note**, there are lots of `NaN`'s. The best solutions will have good ways to handle missing values. For now, we'll ignore them."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"submission_labels.head()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>2008 [YR2008]</th>\n",
" <th>2012 [YR2012]</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>559 </th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>618 </th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>753 </th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1030</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1896</th>\n",
" <td>NaN</td>\n",
" <td>NaN</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>5 rows \u00d7 2 columns</p>\n",
"</div>"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 4,
"text": [
" 2008 [YR2008] 2012 [YR2012]\n",
"559 NaN NaN\n",
"618 NaN NaN\n",
"753 NaN NaN\n",
"1030 NaN NaN\n",
"1896 NaN NaN\n",
"\n",
"[5 rows x 2 columns]"
]
}
],
"prompt_number": 4
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Submission labels are simpler. First observation is we want to predict 2008 and 2012 (and not the years in between). Second observation is that the index values indentify specific rows in the training set. E.g., we want to predict the row in the training set that has the ID 559. Just to look that up:"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"training_data.loc[559]"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 5,
"text": [
"1972 [YR1972] NaN\n",
"1973 [YR1973] NaN\n",
"1974 [YR1974] NaN\n",
"1975 [YR1975] NaN\n",
"1976 [YR1976] NaN\n",
"1977 [YR1977] NaN\n",
"1978 [YR1978] NaN\n",
"1979 [YR1979] NaN\n",
"1980 [YR1980] NaN\n",
"1981 [YR1981] NaN\n",
"1982 [YR1982] NaN\n",
"1983 [YR1983] NaN\n",
"1984 [YR1984] NaN\n",
"1985 [YR1985] NaN\n",
"1986 [YR1986] NaN\n",
"1987 [YR1987] NaN\n",
"1988 [YR1988] NaN\n",
"1989 [YR1989] NaN\n",
"1990 [YR1990] NaN\n",
"1991 [YR1991] 0.048\n",
"1992 [YR1992] 0.049\n",
"1993 [YR1993] 0.049\n",
"1994 [YR1994] 0.049\n",
"1995 [YR1995] 0.049\n",
"1996 [YR1996] 0.084\n",
"1997 [YR1997] 0.118\n",
"1998 [YR1998] 0.152\n",
"1999 [YR1999] 0.187\n",
"2000 [YR2000] 0.221\n",
"2001 [YR2001] 0.256\n",
"2002 [YR2002] 0.291\n",
"2003 [YR2003] 0.325\n",
"2004 [YR2004] 0.36\n",
"2005 [YR2005] 0.395\n",
"2006 [YR2006] 0.43\n",
"2007 [YR2007] 0.465\n",
"Country Name Afghanistan\n",
"Series Code 7.8\n",
"Series Name Ensure environmental sustainability\n",
"Name: 559, dtype: object"
]
}
],
"prompt_number": 5
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"And now we can see that we want to predict the time series for the [Millennium Development Goal 7.8](http://unstats.un.org/unsd/mdg/Host.aspx?Content=Indicators/OfficialList.htm), which is about ensuring environmental sustainability.\n",
"\n",
"We're almost to the fun part. But first let's just write a little utility function to keep us sane. We noticed that the year columns have annoying titles (sorry!). We'll write a little function that will make it easier to grab any column that we want."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"def generate_year_list(start, stop=None):\n",
" \"\"\" \n",
" make a list of column names for specific years\n",
" in the format they appear in the data frame start/stop inclusive\n",
" \"\"\"\n",
" \n",
" if isinstance(start, list):\n",
" data_range = start\n",
" elif stop:\n",
" data_range = range(start, stop+1)\n",
" else:\n",
" data_range = [start]\n",
" \n",
" yrs = []\n",
" \n",
" for yr in data_range:\n",
" yrs.append(\"{0} [YR{0}]\".format(yr))\n",
" \n",
" return yrs\n",
"\n",
"# ========== TEST CASES =======\n",
"# one year\n",
"print generate_year_list(2008)\n",
"\n",
"# start and stop (inclusive)\n",
"print generate_year_list(1985, 1990)\n",
"\n",
"# custom year list\n",
"print generate_year_list([1985, 1990])"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"['2008 [YR2008]']\n",
"['1985 [YR1985]', '1986 [YR1986]', '1987 [YR1987]', '1988 [YR1988]', '1989 [YR1989]', '1990 [YR1990]']\n",
"['1985 [YR1985]', '1990 [YR1990]']\n"
]
}
],
"prompt_number": 6
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Alright, let's start by looking at the time series that we want to predict. In `pandas` it's pretty easy to grab just those rows from the training set since we have the IDs in `submission_labels`. Also, to make our lives easier, we'll drop the info about what these rows are. We can always grab that again from the training data when we want. This way we're just looking at the numeric matrix. "
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"prediction_rows = training_data.loc[submission_labels.index]\n",
"prediction_rows = prediction_rows[generate_year_list(1972, 2007)]\n",
"prediction_rows.head()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>1972 [YR1972]</th>\n",
" <th>1973 [YR1973]</th>\n",
" <th>1974 [YR1974]</th>\n",
" <th>1975 [YR1975]</th>\n",
" <th>1976 [YR1976]</th>\n",
" <th>1977 [YR1977]</th>\n",
" <th>1978 [YR1978]</th>\n",
" <th>1979 [YR1979]</th>\n",
" <th>1980 [YR1980]</th>\n",
" <th>1981 [YR1981]</th>\n",
" <th>1982 [YR1982]</th>\n",
" <th>1983 [YR1983]</th>\n",
" <th>1984 [YR1984]</th>\n",
" <th>1985 [YR1985]</th>\n",
" <th>1986 [YR1986]</th>\n",
" <th>1987 [YR1987]</th>\n",
" <th>1988 [YR1988]</th>\n",
" <th>1989 [YR1989]</th>\n",
" <th>1990 [YR1990]</th>\n",
" <th>1991 [YR1991]</th>\n",
" <th>1992 [YR1992]</th>\n",
" <th>1993 [YR1993]</th>\n",
" <th>1994 [YR1994]</th>\n",
" <th>1995 [YR1995]</th>\n",
" <th>1996 [YR1996]</th>\n",
" <th>1997 [YR1997]</th>\n",
" <th>1998 [YR1998]</th>\n",
" <th>1999 [YR1999]</th>\n",
" <th>2000 [YR2000]</th>\n",
" <th>2001 [YR2001]</th>\n",
" <th>2002 [YR2002]</th>\n",
" <th>2003 [YR2003]</th>\n",
" <th>2004 [YR2004]</th>\n",
" <th>2005 [YR2005]</th>\n",
" <th>2006 [YR2006]</th>\n",
" <th>2007 [YR2007]</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>559 </th>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> 0.0480</td>\n",
" <td> 0.0490</td>\n",
" <td> 0.0490</td>\n",
" <td> 0.0490</td>\n",
" <td> 0.0490</td>\n",
" <td> 0.0840</td>\n",
" <td> 0.1180</td>\n",
" <td> 0.1520</td>\n",
" <td> 0.1870</td>\n",
" <td> 0.2210</td>\n",
" <td> 0.256000</td>\n",
" <td> 0.291000</td>\n",
" <td> 0.325000</td>\n",
" <td> 0.360000</td>\n",
" <td> 0.395000</td>\n",
" <td> 0.430000</td>\n",
" <td> 0.4650</td>\n",
" </tr>\n",
" <tr>\n",
" <th>618 </th>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> 0.0000</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> 0.000047</td>\n",
" <td> 0.000046</td>\n",
" <td> 0.000879</td>\n",
" <td> 0.001058</td>\n",
" <td> 0.012241</td>\n",
" <td> 0.021071</td>\n",
" <td> 0.0190</td>\n",
" </tr>\n",
" <tr>\n",
" <th>753 </th>\n",
" <td> 0.296</td>\n",
" <td> 0.2909</td>\n",
" <td> 0.2852</td>\n",
" <td> 0.2798</td>\n",
" <td> 0.2742</td>\n",
" <td> 0.2683</td>\n",
" <td> 0.2624</td>\n",
" <td> 0.2565</td>\n",
" <td> 0.2503</td>\n",
" <td> 0.2439</td>\n",
" <td> 0.2374</td>\n",
" <td> 0.2304</td>\n",
" <td> 0.2229</td>\n",
" <td> 0.2151</td>\n",
" <td> 0.2071</td>\n",
" <td> 0.1993</td>\n",
" <td> 0.1914</td>\n",
" <td> 0.1836</td>\n",
" <td> 0.1762</td>\n",
" <td> 0.1693</td>\n",
" <td> 0.1627</td>\n",
" <td> 0.1571</td>\n",
" <td> 0.1521</td>\n",
" <td> 0.1479</td>\n",
" <td> 0.1446</td>\n",
" <td> 0.1417</td>\n",
" <td> 0.1391</td>\n",
" <td> 0.1366</td>\n",
" <td> 0.1339</td>\n",
" <td> 0.131000</td>\n",
" <td> 0.127700</td>\n",
" <td> 0.124400</td>\n",
" <td> 0.121000</td>\n",
" <td> 0.117700</td>\n",
" <td> 0.114500</td>\n",
" <td> 0.1115</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1030</th>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> 0.0010</td>\n",
" <td> 0.0010</td>\n",
" <td> 0.0010</td>\n",
" <td> 0.0010</td>\n",
" <td> 0.0010</td>\n",
" <td> 0.0010</td>\n",
" <td> 0.0010</td>\n",
" <td> 0.0010</td>\n",
" <td> 0.0010</td>\n",
" <td> 0.0010</td>\n",
" <td> 0.0010</td>\n",
" <td> 0.001000</td>\n",
" <td> 0.001000</td>\n",
" <td> 0.001000</td>\n",
" <td> 0.001000</td>\n",
" <td> 0.001000</td>\n",
" <td> 0.001000</td>\n",
" <td> 0.0010</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1896</th>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> 0.9640</td>\n",
" <td> 0.9640</td>\n",
" <td> 0.9650</td>\n",
" <td> 0.9650</td>\n",
" <td> 0.9650</td>\n",
" <td> 0.9650</td>\n",
" <td> 0.9650</td>\n",
" <td> 0.964000</td>\n",
" <td> 0.964000</td>\n",
" <td> 0.963000</td>\n",
" <td> 0.963000</td>\n",
" <td> 0.962000</td>\n",
" <td> 0.962000</td>\n",
" <td> 0.9610</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>5 rows \u00d7 36 columns</p>\n",
"</div>"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 7,
"text": [
" 1972 [YR1972] 1973 [YR1973] 1974 [YR1974] 1975 [YR1975] \\\n",
"559 NaN NaN NaN NaN \n",
"618 NaN NaN NaN NaN \n",
"753 0.296 0.2909 0.2852 0.2798 \n",
"1030 NaN NaN NaN NaN \n",
"1896 NaN NaN NaN NaN \n",
"\n",
" 1976 [YR1976] 1977 [YR1977] 1978 [YR1978] 1979 [YR1979] \\\n",
"559 NaN NaN NaN NaN \n",
"618 NaN NaN NaN NaN \n",
"753 0.2742 0.2683 0.2624 0.2565 \n",
"1030 NaN NaN NaN NaN \n",
"1896 NaN NaN NaN NaN \n",
"\n",
" 1980 [YR1980] 1981 [YR1981] 1982 [YR1982] 1983 [YR1983] \\\n",
"559 NaN NaN NaN NaN \n",
"618 NaN NaN NaN NaN \n",
"753 0.2503 0.2439 0.2374 0.2304 \n",
"1030 NaN NaN NaN NaN \n",
"1896 NaN NaN NaN NaN \n",
"\n",
" 1984 [YR1984] 1985 [YR1985] 1986 [YR1986] 1987 [YR1987] \\\n",
"559 NaN NaN NaN NaN \n",
"618 NaN NaN NaN NaN \n",
"753 0.2229 0.2151 0.2071 0.1993 \n",
"1030 NaN NaN NaN NaN \n",
"1896 NaN NaN NaN NaN \n",
"\n",
" 1988 [YR1988] 1989 [YR1989] 1990 [YR1990] 1991 [YR1991] \\\n",
"559 NaN NaN NaN 0.0480 \n",
"618 NaN NaN 0.0000 NaN \n",
"753 0.1914 0.1836 0.1762 0.1693 \n",
"1030 NaN NaN 0.0010 0.0010 \n",
"1896 NaN NaN NaN NaN \n",
"\n",
" 1992 [YR1992] 1993 [YR1993] 1994 [YR1994] 1995 [YR1995] \\\n",
"559 0.0490 0.0490 0.0490 0.0490 \n",
"618 NaN NaN NaN NaN \n",
"753 0.1627 0.1571 0.1521 0.1479 \n",
"1030 0.0010 0.0010 0.0010 0.0010 \n",
"1896 NaN NaN 0.9640 0.9640 \n",
"\n",
" 1996 [YR1996] 1997 [YR1997] 1998 [YR1998] 1999 [YR1999] \\\n",
"559 0.0840 0.1180 0.1520 0.1870 \n",
"618 NaN NaN NaN NaN \n",
"753 0.1446 0.1417 0.1391 0.1366 \n",
"1030 0.0010 0.0010 0.0010 0.0010 \n",
"1896 0.9650 0.9650 0.9650 0.9650 \n",
"\n",
" 2000 [YR2000] 2001 [YR2001] 2002 [YR2002] 2003 [YR2003] \\\n",
"559 0.2210 0.256000 0.291000 0.325000 \n",
"618 NaN 0.000047 0.000046 0.000879 \n",
"753 0.1339 0.131000 0.127700 0.124400 \n",
"1030 0.0010 0.001000 0.001000 0.001000 \n",
"1896 0.9650 0.964000 0.964000 0.963000 \n",
"\n",
" 2004 [YR2004] 2005 [YR2005] 2006 [YR2006] 2007 [YR2007] \n",
"559 0.360000 0.395000 0.430000 0.4650 \n",
"618 0.001058 0.012241 0.021071 0.0190 \n",
"753 0.121000 0.117700 0.114500 0.1115 \n",
"1030 0.001000 0.001000 0.001000 0.0010 \n",
"1896 0.963000 0.962000 0.962000 0.9610 \n",
"\n",
"[5 rows x 36 columns]"
]
}
],
"prompt_number": 7
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Now let's just get a visual sense of what some of these look like. We'll grab 10 of the series at random and plot them below."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# grab a random sample of 10 of the timeseries\n",
"np.random.seed(896)\n",
"rand_rows = np.random.choice(prediction_rows.index.values, size=10)\n",
"\n",
"def plot_rows(data, ids=None, linestyle=\"-\", legend=True):\n",
" # get some colors for the lines\n",
" bmap = pplt.brewer2mpl.get_map('Set3','Qualitative', 10)\n",
" colors = bmap.mpl_colors\n",
" \n",
" if not None == ids:\n",
" get_rows = lambda: enumerate(ids)\n",
" else:\n",
" get_rows = lambda: enumerate(data.index.values)\n",
" \n",
" for i, r in get_rows():\n",
" # get the time series values\n",
" time_data = data.loc[r]\n",
"\n",
" # create an x axis to plot along\n",
" just_years = [y[:4] for y in data.columns]\n",
" X = pd.DatetimeIndex(just_years)\n",
"\n",
" # get time series info for labeling\n",
" country, descrip = training_data[[\"Country Name\", \"Series Name\"]].loc[r]\n",
"\n",
" # plot the series\n",
" plt.plot(X, time_data, c=colors[i],\n",
" label=\"{} - {}\".format(country, descrip), ls=linestyle)\n",
" plt.scatter(X, time_data, alpha=0.8, c=colors[i])\n",
"\n",
" if legend:\n",
" plt.legend(loc=0)\n",
" plt.title(\"Progress Towards Subset of MDGs\")\n",
"\n",
"plot_rows(prediction_rows, ids=rand_rows)\n",
"plt.show()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAzQAAAHpCAYAAABUTXmvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
"text": [
"<matplotlib.figure.Figure at 0x11070e4d0>"
]
}
],
"prompt_number": 8
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"<a id=\"model\"></a>\n",
"\n",
"# Making a simple model\n",
"\n",
"<hr>\n",
"\n",
"We'll start with a super simple model. If we only have data for 2007 and not before, we'll predict that value for 2008 and 2012. If we have more data, we'll do a simple linear regression using just the data points for 2006 and 2007. Our prediction will be that the line which passes through 2006 and 2007 also passes through 2008 and 2012. Hey, maybe you can think of something even more clever to **[move you up the leaderboard](http://www.drivendata.org/competitions/1/leaderboard/)**!"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"def simple_model(series):\n",
" point_2007 = series.iloc[-1]\n",
" point_2006 = series.iloc[-2]\n",
" \n",
" # if just one point, status quo\n",
" if np.isnan(point_2006):\n",
" predictions = np.array([point_2007, point_2007])\n",
" else:\n",
" slope = point_2007 - point_2006\n",
" \n",
" # one year\n",
" pred_2008 = point_2007 + slope\n",
" \n",
" # five years\n",
" pred_2012 = point_2007 + 5*slope\n",
" \n",
" predictions = np.array([pred_2008, pred_2012])\n",
"\n",
" ix = pd.Index(generate_year_list([2008, 2012]))\n",
" return pd.Series(data=predictions, index=ix)\n",
" \n",
"# let's try just these predictions on the first five rows\n",
"test_data = prediction_rows.head()\n",
"test_predictions = test_data.apply(simple_model, axis=1)\n",
"\n",
"# combine the data and the predictions\n",
"test_predictions = test_data.join(test_predictions)\n",
"\n",
"# let's take a look at 2006, 2007, and our predictions\n",
"test_predictions[generate_year_list([2006, 2007, 2008, 2012])]"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>2006 [YR2006]</th>\n",
" <th>2007 [YR2007]</th>\n",
" <th>2008 [YR2008]</th>\n",
" <th>2012 [YR2012]</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>559 </th>\n",
" <td> 0.430000</td>\n",
" <td> 0.4650</td>\n",
" <td> 0.500000</td>\n",
" <td> 0.640000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>618 </th>\n",
" <td> 0.021071</td>\n",
" <td> 0.0190</td>\n",
" <td> 0.016929</td>\n",
" <td> 0.008644</td>\n",
" </tr>\n",
" <tr>\n",
" <th>753 </th>\n",
" <td> 0.114500</td>\n",
" <td> 0.1115</td>\n",
" <td> 0.108500</td>\n",
" <td> 0.096500</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1030</th>\n",
" <td> 0.001000</td>\n",
" <td> 0.0010</td>\n",
" <td> 0.001000</td>\n",
" <td> 0.001000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1896</th>\n",
" <td> 0.962000</td>\n",
" <td> 0.9610</td>\n",
" <td> 0.960000</td>\n",
" <td> 0.956000</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>5 rows \u00d7 4 columns</p>\n",
"</div>"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 9,
"text": [
" 2006 [YR2006] 2007 [YR2007] 2008 [YR2008] 2012 [YR2012]\n",
"559 0.430000 0.4650 0.500000 0.640000\n",
"618 0.021071 0.0190 0.016929 0.008644\n",
"753 0.114500 0.1115 0.108500 0.096500\n",
"1030 0.001000 0.0010 0.001000 0.001000\n",
"1896 0.962000 0.9610 0.960000 0.956000\n",
"\n",
"[5 rows x 4 columns]"
]
}
],
"prompt_number": 9
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"OK, so our dead-simple regression is working. Let's see how it looks when we combine it with the data that we graphed earlier. "
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# make the predictions\n",
"predictions = prediction_rows.loc[rand_rows].apply(simple_model, axis=1)\n",
"\n",
"# plot the data\n",
"plot_rows(prediction_rows, ids=rand_rows)\n",
"\n",
"# plot the predictions\n",
"plot_rows(predictions, linestyle=\"--\", legend=False)\n",
"\n",
"plt.show()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAzQAAAHpCAYAAABUTXmvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nBRoREREREclZ/x8iTeqozPZepAAAAABJRU5ErkJggg==\n",
"text": [
"<matplotlib.figure.Figure at 0x10b2e7390>"
]
}
],
"prompt_number": 10
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We've made some predictions! These don't look too crazy, so let's write out a submission file that we can upload on on the [competition submissions page](http://www.drivendata.org/competitions/1/submissions/). We've put together a little utility function that will write out the submissions in the right format. It will accept the following data structures:\n",
"\n",
" - `pandas.DataFrame` - with index and columns that are already specified properly (as in this notebook)\n",
" - `numpy.array` - with 2 columns (just the predictions) or 3 columns (id, 2008, 2012)\n",
" - `list` - must be a list with two lists as its elements; the first, 2008, the second, 2012\n",
" - `dict` - must have two keys that match the prediction indices ('2008 [YR2008]', '2012 [YR2012]') with values that are a list of predictions"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"def write_submission_file(preds, filename):\n",
" # load the submission labels\n",
" file_format = pd.read_csv(os.path.join(\"data\", \"SubmissionRows.csv\"), index_col=0)\n",
" expected_row_count = file_format.shape[0]\n",
"\n",
" if isinstance(preds, pd.DataFrame):\n",
" # check indices\n",
" assert(preds.index == file_format.index).all(), \\\n",
" \"DataFrame: Prediction indices must match submission format.\"\n",
" \n",
" # check columns\n",
" assert (preds.columns == file_format.columns).all(), \\\n",
" \"DataFrame: Column names must match submission format.\"\n",
" \n",
" final_predictions = preds\n",
" \n",
" elif isinstance(preds, np.ndarray):\n",
" rows, cols = preds.shape\n",
" \n",
" if cols == 3:\n",
" assert (preds[:,0] == file_format.index.values).all(), \\\n",
" \"Numpy Array: First column must be indices.\"\n",
" \n",
" # now we know the indices are cool, ditch them\n",
" preds = preds[:,1:]\n",
" \n",
" assert rows == expected_row_count, \\\n",
" \"Numpy Array: The predictions must have the right number of rows.\"\n",
" \n",
" # put the predictions into the dataframe\n",
" final_predictions = file_format.copy()\n",
" final_predictions[generate_year_list([2008, 2012])] = preds\n",
" \n",
" elif isinstance(preds, list):\n",
" assert len(preds) == 2, \\\n",
" \"list: Predictions must be a list containing two lists\"\n",
" assert len(preds[0]) == expected_row_count, \\\n",
" \"list: There must be the right number of predictions in the first list.\"\n",
" assert len(preds[1]) == expected_row_count, \\\n",
" \"list: There must be the right number of predictions in the second list.\"\n",
" \n",
" # write the predictions\n",
" final_predictions = file_format.copy()\n",
" final_predictions[generate_year_list(2008)] = np.array(preds[0], dtype=np.float64).reshape(-1, 1)\n",
" final_predictions[generate_year_list(2012)] = np.array(preds[1], dtype=np.float64).reshape(-1, 1)\n",
" \n",
" elif isinstance(preds, dict):\n",
" assert preds.keys() == generate_year_list([2008, 2012]), \\\n",
" \"dict: keys must be properly formatted\"\n",
" assert len(preds[generate_year_list(2008)[0]]) == expected_row_count, \\\n",
" \"dict: length of value for 2008 must match the number of predictions\"\n",
" assert len(preds[generate_year_list(2012)[0]]) == expected_row_count, \\\n",
" \"dict: length of value for 2012 must match the number of predictions\"\n",
" \n",
" # create dataframe from dictionary\n",
" final_predictions = pd.DataFrame(preds, index=file_format.index)\n",
"\n",
" final_predictions.to_csv(filename)\n",
" \n",
"simple_predictions = prediction_rows.apply(simple_model, axis=1)\n",
"write_submission_file(simple_predictions, \"Getting Started Benchmark.csv\")"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 11
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"<a id=\"correlations\"></a>\n",
"\n",
"# Starting to think about correlations\n",
"\n",
"<hr>\n",
"\n",
"We won't make any prediction based on this section of the notebook (that's up to you!), but one hypothesis you might have is that certain other timeseries are correlated with the ones we want to predict.\n",
"\n",
"As an example, say we are looking at GDP and percent of population that makes under a dollar a day (PLDD). You may think that an increase in GDP during 2006-2007 would entail a decrease PLDD during 2007-2008. Similarly a decrease during 2006-2007 in GDP would result in an increase in PLDD. \n",
"\n",
"We might call this a lagged correlation. As an experiment, we'll pick one country, Kenya, and make a correlation matrix between all of the indicators."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"kenya_data = training_data[training_data[\"Country Name\"] == 'Kenya']\n",
"kenya_values = kenya_data[generate_year_list(1972, 2007)].values\n",
"\n",
"# get the total number of time series we have for Kenya\n",
"nseries = kenya_values.shape[0]\n",
"\n",
"# -1 as default\n",
"lag_corr_mat = np.ones([nseries, nseries], dtype=np.float64)*-1\n",
"\n",
"# create a matrix to hold our lagged correlations\n",
"for i in range(nseries):\n",
" for j in range(nseries):\n",
" # skip comparing a series with itself\n",
" if i!=j:\n",
" # get original (1972-2006) and shifted (1973-2007)\n",
" original = kenya_values[i,1:]\n",
" shifted = kenya_values[j,:-1]\n",
"\n",
" # for just the indices where neither is nan\n",
" non_nan_mask = (~np.isnan(original) & ~np.isnan(shifted))\n",
"\n",
" # if we have at least 2 data points\n",
" if non_nan_mask.sum() >= 2:\n",
" lag_corr_mat[i,j] = np.correlate(original[non_nan_mask], shifted[non_nan_mask])"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 12
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Finally, let's see what the most correlated lagged time series is for one of the rows we have to predict. I happen to know that the row with the index 131042 is one we need to predict."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# let's look at one of the indicators we are suppoed to predict\n",
"to_predict_ix = 131042 \n",
"\n",
"# first, we get the index of that row in the correlation matrix\n",
"i = np.where(kenya_data.index.values == to_predict_ix)[0][0]\n",
"\n",
"# then, we see which value in the matrix is the largest for that row\n",
"j_max = np.argmax(lag_corr_mat[i,:])\n",
"\n",
"# finally, let's see what these correspond to\n",
"max_corr_ix = kenya_data.index.values[j_max]\n",
"\n",
"# now write out what we've found\n",
"fmt_string = \"In Kenya, the progress of '{}' is \"\\\n",
" \"most correlated with a change in '{}' during the year before.\"\n",
" \n",
"print fmt_string.format(kenya_data[\"Series Name\"][to_predict_ix], kenya_data[\"Series Name\"][max_corr_ix])"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"In Kenya, the progress of 'Achieve universal primary education' is most correlated with a change in 'Gross national expenditure (current LCU)' during the year before.\n"
]
}
],
"prompt_number": 13
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Hey, that's actually plausible! Awesome!\n",
"\n",
"That's all for now. It's your turn to uncover some of these interesting structures yourself and [gain some notoreity](http://www.drivendata.org/competitions/1/) along the way."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"If you're looking for a place to start, Hyndman's open textbook [\"Forecasting: principles and practice\"](https://www.otexts.org/fpp/) will definitely give you some ideas. Code snippets are in R, but translatable to Python.\n",
"\n",
"# Use this notebook as a jumping off point and [make some submissions](http://www.drivendata.org/competitions/1/). Good luck!"
]
}
],
"metadata": {}
}
]
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment