jkibele/StatsBookChpt5.ipynb

## StatsBookChpt5.ipynb
{
 "metadata": {
  "name": "StatsBookChpt5"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "heading",
     "level": 3,
     "metadata": {},
     "source": "What's this all about?"
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "I'm working on getting a better grasp of some statistical concepts that I need for my research. I'm using [Experimental Design and Data Analysis for Biologists](http://books.google.co.nz/books/about/Experimental_Design_and_Data_Analysis_fo.html?id=VtU3-y7LaLYC&redir_esc=y) (referred to as 'the book' from here on). At the same time, I want to learn how to perform the analyses I'm learning about using open source Python. This is the [IPython Notebook](http://ipython.org/notebook.html) I put together for some of the topics in Chapter 5 of the book. If you have a bit of familiarity with Python, this should help you figure out some statistics. ...maybe..."
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "Import the Python libraries I'm planning to use."
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "from scipy import stats\nimport pandas as pd",
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 28
    },
    {
     "cell_type": "heading",
     "level": 3,
     "metadata": {},
     "source": "Replicating figure 5.3"
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "Here's how to create a plot kind of like figure 5.3 from the book. I'm just focusing on how to recreate this type of plot for the moment so I'll just produce some random data to mess with."
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "x = np.arange(1,101,0.5)\nrnds = np.random.rand(200) + np.random.randint(-20,20,200)\ny = x + rnds\nxrand = x + np.random.rand(200) + np.random.randint(-30,30,200)",
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 29
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "I'm going to use matplotlib's [GridSpec](http://matplotlib.org/users/gridspec.html) to layout the plot."
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "import matplotlib.gridspec as gspec\ngs = gspec.GridSpec(2,2,width_ratios=[6,1],height_ratios=[1,4])\n\nsubplot(gs[0])\nboxplot(xrand,vert=False)\naxis('off')\nsubplot(gs[2])\nscatter(xrand,y)\n# scaled would be nice (it makes the plot more square)\n# but it only changes the axses of the scatter so that \n# the box plots don't line up anymore. I'm too lazy to\n# figure out a fix right now.\n#axis('scaled') \nsubplot(gs[3])\nboxplot(y)\naxis('off')",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 30,
       "text": "(0.5, 1.5, -20.0, 120.0)"
      },
      {
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD5CAYAAAA3Os7hAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEAichP8HtyJzhrYXjREAAAAASUVORK5CYII=\n"
      }
     ],
     "prompt_number": 30
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "# polyfit will give the slope and intercept\nm,b = polyfit(xrand,y,1)\nyp = polyval([m,b],xrand)\nscatter(xrand,y)\nplot(xrand,yp)\nprint \"m=%.3f, b=%.3f\" % (m,b)",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "m=0.750, b=11.029\n"
      },
      {
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAD9CAYAAABQvqc9AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n"
      }
     ],
     "prompt_number": 31
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "Using `stats.linregress` is better because it returns more information. Here's the list of returns from [the SciPy Documentation](http://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.linregress.html): \n\n```\nslope : float\n    slope of the regression line\nintercept : float\n    intercept of the regression line\nr-value : float\n    correlation coefficient\np-value : float\n    two-sided p-value for a hypothesis test whose null hypothesis is\n    that the slope is zero.\nstderr : float\n    Standard error of the estimate\n```"
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "stats.linregress(xrand,y)",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 32,
       "text": "(0.7498010309757458,\n 11.029471497053791,\n 0.80676101689385427,\n 3.9967506042857742e-47,\n 0.039027098172076814)"
      }
     ],
     "prompt_number": 32
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "So $r^2$ would be:"
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "stats.linregress(xrand,y)[2]**2",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 33,
       "text": "0.65086333837960586"
      }
     ],
     "prompt_number": 33
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "Here's another way to get the correlation coefficient."
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "corrcoef(xrand,y)",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 34,
       "text": "array([[ 1.        ,  0.80676102],\n       [ 0.80676102,  1.        ]])"
      }
     ],
     "prompt_number": 34
    },
    {
     "cell_type": "heading",
     "level": 3,
     "metadata": {},
     "source": "Box 5.1"
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "Now we'll try to do Box 5.1 from [Experimental Design and Data Analysis for Biologists](http://books.google.co.nz/books/about/Experimental_Design_and_Data_Analysis_fo.html?id=VtU3-y7LaLYC&redir_esc=y). I'm going to use the actual data used in the book (it's available online) and demonstrate how you'd reproduce the plots in the book (more or less) using Python."
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "# get the data\nxmas_island = pd.read_csv('http://www.zoology.unimelb.edu.au/qkstats/chpt5/green.csv')\n# put the data from the two sites in their own data frames for ease of use\nDS = xmas_island[xmas_island.SITE=='DS']\nLS = xmas_island[xmas_island.SITE=='LS']\n# have a look at the LS data frame\nLS.head() # the head method will display the first five rows",
     "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>SITE</th>\n      <th>QUADNUM</th>\n      <th>TOTMASS</th>\n      <th>BURROWS</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>8 </th>\n      <td> LS</td>\n      <td> 1</td>\n      <td> 4.36</td>\n      <td> 38</td>\n    </tr>\n    <tr>\n      <th>9 </th>\n      <td> LS</td>\n      <td> 2</td>\n      <td> 4.01</td>\n      <td> 37</td>\n    </tr>\n    <tr>\n      <th>10</th>\n      <td> LS</td>\n      <td> 3</td>\n      <td> 3.33</td>\n      <td> 27</td>\n    </tr>\n    <tr>\n      <th>11</th>\n      <td> LS</td>\n      <td> 4</td>\n      <td> 2.63</td>\n      <td> 18</td>\n    </tr>\n    <tr>\n      <th>12</th>\n      <td> LS</td>\n      <td> 5</td>\n      <td> 4.46</td>\n      <td> 41</td>\n    </tr>\n  </tbody>\n</table>\n</div>",
       "output_type": "pyout",
       "prompt_number": 35,
       "text": "   SITE  QUADNUM  TOTMASS  BURROWS\n8    LS        1     4.36       38\n9    LS        2     4.01       37\n10   LS        3     3.33       27\n11   LS        4     2.63       18\n12   LS        5     4.46       41"
      }
     ],
     "prompt_number": 35
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "Pearson, Spearman, and Kendall statistics and $P$ values are super easy to calculate:"
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "# Pearson assumes bivariate normal distribution\nls_pearson_r, ls_pearson_p = stats.pearsonr(LS.TOTMASS, LS.BURROWS)\n# The next two are rank based so don't assume normal distribution\n# They should work for any monotonic relationship\nls_spearman_r, ls_spearman_p = stats.spearmanr(LS.TOTMASS, LS.BURROWS)\nls_kendall_tau, ls_kendall_p = stats.kendalltau(LS.TOTMASS, LS.BURROWS)",
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 36
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "I'll stick the values in a pandas data frame so they display in a nice little table thing:"
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "ls_d = { 'Statistic': pd.Series([ls_pearson_r,ls_spearman_r,ls_kendall_tau], index=['Pearson','Spearman','Kendall']),\n         'P Value'  : pd.Series([ls_pearson_p,ls_spearman_p,ls_kendall_p], index=['Pearson','Spearman','Kendall']) }\nls_df = pd.DataFrame( ls_d )\nls_df",
     "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>P Value</th>\n      <th>Statistic</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>Pearson</th>\n      <td> 0.000735</td>\n      <td> 0.881955</td>\n    </tr>\n    <tr>\n      <th>Spearman</th>\n      <td> 0.001791</td>\n      <td> 0.851068</td>\n    </tr>\n    <tr>\n      <th>Kendall</th>\n      <td> 0.003798</td>\n      <td> 0.719147</td>\n    </tr>\n  </tbody>\n</table>\n</div>",
       "output_type": "pyout",
       "prompt_number": 37,
       "text": "           P Value  Statistic\nPearson   0.000735   0.881955\nSpearman  0.001791   0.851068\nKendall   0.003798   0.719147"
      }
     ],
     "prompt_number": 37
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "The Pearson correlation is the same thing as the correlation coefficient $r$, right? It looks like it. (remember, $r$ is the 3rd value spit out by linregress)"
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "stats.linregress(LS.TOTMASS,LS.BURROWS)",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 38,
       "text": "(9.4152096922052131,\n -3.1495512013031899,\n 0.88195494383269679,\n 0.00073499375761430276,\n 1.7789638248707702)"
      }
     ],
     "prompt_number": 38
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "Now I'm going to just more or less reproduce figure 5.3 from the book. I may want to do something similar for my own work so it seems worth doing."
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "figure(figsize=[10,5])\ngs = gspec.GridSpec(2,4,width_ratios=[3,.5,3,.5],height_ratios=[1,4])\n\ngs0 = subplot(gs[0])\nboxplot(LS.TOTMASS.values,vert=False)\naxis('off')\nsubplot(gs[4])\nscat_ls = scatter(LS.TOTMASS.values,LS.BURROWS.values)\n# That didn't work and I can't be bothered to figure out why right now.\n#scat_ls.axes.text(scat_ls.axes.get_ylim()[1]-1, scat_ls.axes.get_xlim()[0]+0.5, 'LS')\ngs5 = subplot(gs[5])\nboxplot(LS.BURROWS.values)\n# The axes weren't staying lined up so I'm going to explicitly size them\ngs0.axes.set_xlim( scat_ls.axes.get_xlim() )\ngs5.axes.set_ylim( scat_ls.axes.get_ylim() )\naxis('off')\n\ngs2 = subplot(gs[2])\nboxplot(DS.TOTMASS.values,vert=False)\naxis('off')\nsubplot(gs[6])\nscat = scatter(DS.TOTMASS.values,DS.BURROWS.values)\ngs7 = subplot(gs[7])\nbpgs7 = boxplot(DS.BURROWS.values)\ngs2.axes.set_xlim( scat.axes.get_xlim() )\ngs7.axes.set_ylim( scat.axes.get_ylim() )\naxis('off')",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 39,
       "text": "(0.5, 1.5, 20.0, 90.0)"
      },
      {
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAksAAAExCAYAAABs7OY4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n"
      }
     ],
     "prompt_number": 39
    },
    {
     "cell_type": "heading",
     "level": 3,
     "metadata": {},
     "source": "Replicating Figure 5.17"
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "# get the data into a pandas data frame\npeake = pd.read_csv('http://www.zoology.unimelb.edu.au/qkstats/chpt5/peake.csv')",
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 40
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "Take a look at some of the data to see what's there. (using the HTML method from IPython just makes the formating a bit nicer)"
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "from IPython.display import HTML\nHTML( peake.head().to_html() )",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": "<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>AREA</th>\n      <th>SPECIES</th>\n      <th>INDIV</th>\n      <th>LAREA</th>\n      <th>LSPECIES</th>\n      <th>LINDIV</th>\n      <th>RESID1</th>\n      <th>PREDICT1</th>\n      <th>RESID2</th>\n      <th>PREDICT2</th>\n      <th>RESID3</th>\n      <th>PREDICT3</th>\n      <th>RESID4</th>\n      <th>PREDICT4</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>  516.00</td>\n      <td>  3</td>\n      <td>  18</td>\n      <td> 2.712650</td>\n      <td> 0.477121</td>\n      <td> 1.255273</td>\n      <td>-7.196363</td>\n      <td> 10.196363</td>\n      <td>-0.296087</td>\n      <td> 0.773208</td>\n      <td>-166.608318</td>\n      <td> 184.608318</td>\n      <td>-0.433547</td>\n      <td> 1.688820</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>  469.06</td>\n      <td>  7</td>\n      <td>  60</td>\n      <td> 2.671228</td>\n      <td> 0.845098</td>\n      <td> 1.778151</td>\n      <td>-3.165416</td>\n      <td> 10.165416</td>\n      <td> 0.087871</td>\n      <td> 0.757227</td>\n      <td>-122.918638</td>\n      <td> 182.918638</td>\n      <td> 0.123915</td>\n      <td> 1.654237</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>  462.25</td>\n      <td>  6</td>\n      <td>  57</td>\n      <td> 2.664877</td>\n      <td> 0.778151</td>\n      <td> 1.755875</td>\n      <td>-4.160926</td>\n      <td> 10.160926</td>\n      <td> 0.023375</td>\n      <td> 0.754776</td>\n      <td>-125.673502</td>\n      <td> 182.673502</td>\n      <td> 0.106941</td>\n      <td> 1.648934</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>  938.60</td>\n      <td>  8</td>\n      <td> 100</td>\n      <td> 2.972481</td>\n      <td> 0.903090</td>\n      <td> 2.000000</td>\n      <td>-2.474981</td>\n      <td> 10.474981</td>\n      <td> 0.029630</td>\n      <td> 0.873460</td>\n      <td> -99.820473</td>\n      <td> 199.820473</td>\n      <td> 0.094243</td>\n      <td> 1.905757</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td> 1357.15</td>\n      <td> 10</td>\n      <td>  48</td>\n      <td> 3.132628</td>\n      <td> 1.000000</td>\n      <td> 1.681241</td>\n      <td>-0.750929</td>\n      <td> 10.750929</td>\n      <td> 0.064750</td>\n      <td> 0.935250</td>\n      <td>-166.886842</td>\n      <td> 214.886842</td>\n      <td>-0.358225</td>\n      <td> 2.039466</td>\n    </tr>\n  </tbody>\n</table>",
       "output_type": "pyout",
       "prompt_number": 41,
       "text": "<IPython.core.display.HTML at 0xbc35c6c>"
      }
     ],
     "prompt_number": 41
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "The full data set has way more columns than I need because I want to calculate that stuff rather than just plot it so I'll make a data frame that has just the data I want."
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "arsp = pd.DataFrame({\"Area\": peake.AREA, \"Species\": peake.SPECIES, \"Indiv\": peake.INDIV} )\narsp.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>Area</th>\n      <th>Indiv</th>\n      <th>Species</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>  516.00</td>\n      <td>  18</td>\n      <td>  3</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>  469.06</td>\n      <td>  60</td>\n      <td>  7</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>  462.25</td>\n      <td>  57</td>\n      <td>  6</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>  938.60</td>\n      <td> 100</td>\n      <td>  8</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td> 1357.15</td>\n      <td>  48</td>\n      <td> 10</td>\n    </tr>\n  </tbody>\n</table>\n</div>",
       "output_type": "pyout",
       "prompt_number": 42,
       "text": "      Area  Indiv  Species\n0   516.00     18        3\n1   469.06     60        7\n2   462.25     57        6\n3   938.60    100        8\n4  1357.15     48       10"
      }
     ],
     "prompt_number": 42
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "In order to get the residuals, it looks like I have to use numpy's [polyfit](http://docs.scipy.org/doc/numpy/reference/generated/numpy.polyfit.html) method with `full=True`."
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "np.polyfit(arsp.Area, arsp.Species, 1, full=True)",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 43,
       "text": "(array([  6.59295373e-04,   9.85616668e+00]),\n array([ 325.0644671]),\n 2,\n array([ 5.22007139,  1.42910565]),\n 5.5511151231257827e-15)"
      }
     ],
     "prompt_number": 43
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "Well, crap. That second array is supposed to be the residuals but there's just a single value there. That's not what I was expecting. I probably misread the documentation or something. Fortunately, there seems to be another way using the [statsmodels](http://statsmodels.sourceforge.net/) package. I'm going to be following [this](http://mpastell.com/2013/04/19/python_regression/) example except that I'm using statsmodels version 0.4.3 so I won't be using the $R$ style api. There's a good IPython notebook demo [here](http://nbviewer.ipython.org/urls/github.com/weecology/progbio/raw/master/ipynbs/statistics.ipynb)."
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "import statsmodels.api as sm\nsm.version.version",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 44,
       "text": "'0.5.0.dev-49d386f'"
      }
     ],
     "prompt_number": 44
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "The example in the [documentation](http://statsmodels.sourceforge.net/stable/generated/statsmodels.regression.linear_model.OLS.html#statsmodels.regression.linear_model.OLS) shows \"`model=sm.OLS(Y,X)`\" I'd sort of expect `X` first, then `Y` but whatever. According to `In [8]` of [this](http://nbviewer.ipython.org/urls/github.com/weecology/progbio/raw/master/ipynbs/statistics.ipynb), I have to add a column of ones to allow the calculation of the intercept. I don't quite grasp why but I guess it's forcing through the origin if I don't? I need to dig through the statsmodels documentation a bit more."
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "X = sm.add_constant( arsp.Area, prepend=True )\nmodel = sm.OLS(arsp.Species, X)\nresults = model.fit()",
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 45
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "`results` is an object with a crap-ton of attributes and methods. Here's the [documentation for it](http://statsmodels.sourceforge.net/stable/generated/statsmodels.regression.linear_model.RegressionResults.html#statsmodels.regression.linear_model.RegressionResults). `results.resid` will return a pandas data series with all of our residuals."
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "results.resid",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 46,
       "text": "0    -7.196363\n1    -3.165416\n2    -4.160926\n3    -2.474981\n4    -0.750929\n5    -2.025533\n6    -0.967745\n7    -0.033859\n8     4.106564\n9    -3.480241\n10    0.226557\n11    1.208861\n12   -1.141363\n13    1.209402\n14    6.523814\n15    7.214803\n16    0.437242\n17    4.111320\n18    5.463147\n19    5.020502\n20   -3.907719\n21    0.759630\n22   -1.149149\n23   -2.752100\n24   -3.075518\ndtype: float64"
      }
     ],
     "prompt_number": 46
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "Now I'm going to try to come up with a function for making plots like Figure 5.17 from the book."
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "def fancy_plot(xarr,yarr,xlabel,ylabel):\n    # generate all the OLS crap\n    X = sm.add_constant( xarr, prepend=True )\n    mod = sm.OLS(yarr, X)\n    res = mod.fit()\n    \n    # generate predicted values for residual plot\n    # first get slope and intercept from results\n    intercept, slope = res.params\n    # define a function based on my regression\n    f = lambda x: intercept + slope * x\n    # generate predicted values for my values of X\n    predicted = f( xarr )\n    \n    # produce smoother line\n    # note that the x and y values are supplied in \n    # reversed order from what I'd expect.\n    smoo = sm.nonparametric.lowess(yarr,xarr,frac=0.5,return_sorted=False)\n    # FYI, that works in statsmodels version 0.5+ but not in 0.4.3.\n    \n    figure(figsize=[15,8])\n    gs = gspec.GridSpec(2,3,width_ratios=[4.5,1,4.5],height_ratios=[1,4])\n    \n    gs0 = subplot(gs[0])\n    boxplot(xarr,vert=False)\n    axis('off')\n    subplot(gs[3])\n    scat = scatter(xarr,yarr,marker='o',facecolors='none',s=50)\n    scat.axes.plot(xarr,smoo)\n    scat.axes.set_ylabel(ylabel) #('Number of Species')\n    scat.axes.set_xlabel(xlabel) #('Clump Area ($m^2$)')\n    gs5 = subplot(gs[4])\n    boxplot(yarr)\n    # The axes weren't staying lined up so I'm going to explicitly size them\n    gs0.axes.set_xlim( scat.axes.get_xlim() )\n    gs5.axes.set_ylim( scat.axes.get_ylim() )\n    axis('off')\n    \n    subplot(gs[5])\n    smoo2 = sm.nonparametric.lowess(res.resid,predicted,frac=0.5,return_sorted=False)\n    scat_resid = scatter(predicted,res.resid,marker='o',facecolors='none',s=50)\n    scat_resid.axes.plot(predicted,smoo2)\n    scat_resid.axes.set_ylabel('Residual')\n    scat_resid.axes.set_xlabel('Predicted Value')",
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 105
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "# see if it works\nfancy_plot(arsp.Area.values, arsp.Species.values, 'Clump Area ($m^2$)', 'Number of Species')",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAA3cAAAHtCAYAAAC6UYcaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n+7/24IAEAAAAQND/1/0IFQAAYCBR83oaGzyJuwAAAABJRU5ErkJggg==\n"
      }
     ],
     "prompt_number": 106
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "Awwwww yeah. That worked but we can see that the fit is pretty crap. `results.summary()` will confirm that."
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "results.summary()",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": "<table class=\"simpletable\">\n<caption>OLS Regression Results</caption>\n<tr>\n  <th>Dep. Variable:</th>         <td>Species</td>     <th>  R-squared:         </th> <td>   0.687</td>\n</tr>\n<tr>\n  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.673</td>\n</tr>\n<tr>\n  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   50.44</td>\n</tr>\n<tr>\n  <th>Date:</th>             <td>Thu, 12 Sep 2013</td> <th>  Prob (F-statistic):</th> <td>3.10e-07</td>\n</tr>\n<tr>\n  <th>Time:</th>                 <td>10:38:57</td>     <th>  Log-Likelihood:    </th> <td> -67.538</td>\n</tr>\n<tr>\n  <th>No. Observations:</th>      <td>    25</td>      <th>  AIC:               </th> <td>   139.1</td>\n</tr>\n<tr>\n  <th>Df Residuals:</th>          <td>    23</td>      <th>  BIC:               </th> <td>   141.5</td>\n</tr>\n<tr>\n  <th>Df Model:</th>              <td>     1</td>      <th>                     </th>     <td> </td>   \n</tr>\n</table>\n<table class=\"simpletable\">\n<tr>\n    <td></td>       <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th> <th>[95.0% Conf. Int.]</th> \n</tr>\n<tr>\n  <th>const</th> <td>    9.8562</td> <td>    1.044</td> <td>    9.441</td> <td> 0.000</td> <td>    7.697    12.016</td>\n</tr>\n<tr>\n  <th>Area</th>  <td>    0.0007</td> <td> 9.28e-05</td> <td>    7.102</td> <td> 0.000</td> <td>    0.000     0.001</td>\n</tr>\n</table>\n<table class=\"simpletable\">\n<tr>\n  <th>Omnibus:</th>       <td> 0.886</td> <th>  Durbin-Watson:     </th> <td>   1.000</td>\n</tr>\n<tr>\n  <th>Prob(Omnibus):</th> <td> 0.642</td> <th>  Jarque-Bera (JB):  </th> <td>   0.893</td>\n</tr>\n<tr>\n  <th>Skew:</th>          <td> 0.367</td> <th>  Prob(JB):          </th> <td>   0.640</td>\n</tr>\n<tr>\n  <th>Kurtosis:</th>      <td> 2.434</td> <th>  Cond. No.          </th> <td>1.56e+04</td>\n</tr>\n</table>",
       "output_type": "pyout",
       "prompt_number": 49,
       "text": "<class 'statsmodels.iolib.summary.Summary'>\n\"\"\"\n                            OLS Regression Results                            \n==============================================================================\nDep. Variable:                Species   R-squared:                       0.687\nModel:                            OLS   Adj. R-squared:                  0.673\nMethod:                 Least Squares   F-statistic:                     50.44\nDate:                Thu, 12 Sep 2013   Prob (F-statistic):           3.10e-07\nTime:                        10:38:57   Log-Likelihood:                -67.538\nNo. Observations:                  25   AIC:                             139.1\nDf Residuals:                      23   BIC:                             141.5\nDf Model:                           1                                         \n==============================================================================\n                 coef    std err          t      P>|t|      [95.0% Conf. Int.]\n------------------------------------------------------------------------------\nconst          9.8562      1.044      9.441      0.000         7.697    12.016\nArea           0.0007   9.28e-05      7.102      0.000         0.000     0.001\n==============================================================================\nOmnibus:                        0.886   Durbin-Watson:                   1.000\nProb(Omnibus):                  0.642   Jarque-Bera (JB):                0.893\nSkew:                           0.367   Prob(JB):                        0.640\nKurtosis:                       2.434   Cond. No.                     1.56e+04\n==============================================================================\n\nWarnings:\n[1] The condition number is large, 1.56e+04. This might indicate that there are\nstrong multicollinearity or other numerical problems.\n\"\"\""
      }
     ],
     "prompt_number": 49
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "Now we'll do the same thing with $\\log_{10}$ transformed data."
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "fancy_plot(log10(arsp.Area.values), log10(arsp.Species.values), 'Log Clump Area', 'Log Number of Species')",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAA3wAAAHnCAYAAADjKl+LAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nB0lEQVSn/j8F0vExIhmgUAAAAABJRU5ErkJggg==\n"
      }
     ],
     "prompt_number": 107
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "Well do it again (not so efficient in terms of computing but this only takes a second) with $\\log_{10}$ values to see the summary table."
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "Xlog = sm.add_constant( log10(arsp.Area), prepend=True )\nmodel_log = sm.OLS(log10(arsp.Species), Xlog)\nresults_log = model_log.fit()\nresults_log.summary()",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": "<table class=\"simpletable\">\n<caption>OLS Regression Results</caption>\n<tr>\n  <th>Dep. Variable:</th>         <td>Species</td>     <th>  R-squared:         </th> <td>   0.819</td>\n</tr>\n<tr>\n  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.812</td>\n</tr>\n<tr>\n  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   104.4</td>\n</tr>\n<tr>\n  <th>Date:</th>             <td>Thu, 12 Sep 2013</td> <th>  Prob (F-statistic):</th> <td>5.11e-10</td>\n</tr>\n<tr>\n  <th>Time:</th>                 <td>10:38:57</td>     <th>  Log-Likelihood:    </th> <td>  23.332</td>\n</tr>\n<tr>\n  <th>No. Observations:</th>      <td>    25</td>      <th>  AIC:               </th> <td>  -42.66</td>\n</tr>\n<tr>\n  <th>Df Residuals:</th>          <td>    23</td>      <th>  BIC:               </th> <td>  -40.23</td>\n</tr>\n<tr>\n  <th>Df Model:</th>              <td>     1</td>      <th>                     </th>     <td> </td>   \n</tr>\n</table>\n<table class=\"simpletable\">\n<tr>\n    <td></td>       <th>coef</th>     <th>std err</th>      <th>t</th>      <th>P>|t|</th> <th>[95.0% Conf. Int.]</th> \n</tr>\n<tr>\n  <th>const</th> <td>   -0.2734</td> <td>    0.138</td> <td>   -1.975</td> <td> 0.060</td> <td>   -0.560     0.013</td>\n</tr>\n<tr>\n  <th>Area</th>  <td>    0.3858</td> <td>    0.038</td> <td>   10.215</td> <td> 0.000</td> <td>    0.308     0.464</td>\n</tr>\n</table>\n<table class=\"simpletable\">\n<tr>\n  <th>Omnibus:</th>       <td>10.887</td> <th>  Durbin-Watson:     </th> <td>   1.781</td>\n</tr>\n<tr>\n  <th>Prob(Omnibus):</th> <td> 0.004</td> <th>  Jarque-Bera (JB):  </th> <td>   9.296</td>\n</tr>\n<tr>\n  <th>Skew:</th>          <td>-1.196</td> <th>  Prob(JB):          </th> <td> 0.00958</td>\n</tr>\n<tr>\n  <th>Kurtosis:</th>      <td> 4.790</td> <th>  Cond. No.          </th> <td>    27.4</td>\n</tr>\n</table>",
       "output_type": "pyout",
       "prompt_number": 51,
       "text": "<class 'statsmodels.iolib.summary.Summary'>\n\"\"\"\n                            OLS Regression Results                            \n==============================================================================\nDep. Variable:                Species   R-squared:                       0.819\nModel:                            OLS   Adj. R-squared:                  0.812\nMethod:                 Least Squares   F-statistic:                     104.4\nDate:                Thu, 12 Sep 2013   Prob (F-statistic):           5.11e-10\nTime:                        10:38:57   Log-Likelihood:                 23.332\nNo. Observations:                  25   AIC:                            -42.66\nDf Residuals:                      23   BIC:                            -40.23\nDf Model:                           1                                         \n==============================================================================\n                 coef    std err          t      P>|t|      [95.0% Conf. Int.]\n------------------------------------------------------------------------------\nconst         -0.2734      0.138     -1.975      0.060        -0.560     0.013\nArea           0.3858      0.038     10.215      0.000         0.308     0.464\n==============================================================================\nOmnibus:                       10.887   Durbin-Watson:                   1.781\nProb(Omnibus):                  0.004   Jarque-Bera (JB):                9.296\nSkew:                          -1.196   Prob(JB):                      0.00958\nKurtosis:                       4.790   Cond. No.                         27.4\n==============================================================================\n\"\"\""
      }
     ],
     "prompt_number": 51
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "Right. So. I'm a bit confused as to why my summary is giving the **const** value as -0.2734. I think that's supposed to be the intercept and box 5.4 of the book lists that as being 1.270. Oh well, I can plot the crap out of things now at least.\n\nI tried the same thing in $R$ and got the same values so I'm not sure what's going on. There's another way to do it in python as well:"
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": "slope, intercept, r_value, p_value, std_err = stats.linregress(log10(arsp.Area.values),log10(arsp.Species.values))\nprint \"slope=%.3f, intercept=%.3f\" % (slope,intercept)",
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": "slope=0.386, intercept=-0.273\n"
      }
     ],
     "prompt_number": 52
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": "Well there are the same values again so I don't know what the deal is. At any rate, the main point of this exercise was for me to be able to produce those plots and the lowess smoother stuff."
    }
   ],
   "metadata": {}
  }
 ]
}