cleemesser/simple-bootstrap.ipynb

## simple-bootstrap.ipynb
{
 "metadata": {
  "name": "simple-bootstrap"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "A basic demonstration of what I'm talking about when I say that you can use the bootstrap to compare mean values between groups of different sizes. \n",
      "\n",
      "The basic idea, is that you can see what the roll of chance was might play given that the smaller group is likely to have much more extreme values than the larger group. Here I will imagine that there are two groups of animals which were each scanned with resultant activity recorded in a certain region. Group A has 5 members and group B has 16 members."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "I'm going to do this in the ipython notebook with pylab loaded. First I will import some useful packages for data analysis:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import pandas as pd\n",
      "import random\n",
      "import scikits.bootstrap as bootstrap  # this provides code to calculate confidence intervals\n",
      "from pandas.tools.plotting import bootstrap_plot"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 1
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Next I will generate the two groups. Group A has five members chosen randomly from 0 to 10.0. Group B has 16 members chosen randomly between 3 and 13. A and B will by numerical arrays."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "A = 10*random_sample(5)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 2
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "B = 10 * random_sample(16) + 3.0"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 3
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "I will also insert A and B into data frames called fA and fB. These are useful for data exploration."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "fA = pd.DataFrame(A, columns=['activity'])\n",
      "fB = pd.DataFrame(B, columns=['activity'])"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 4
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Let's look at the data values"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "fA"
     ],
     "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>activity</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
        "      <td> 3.655182</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1</th>\n",
        "      <td> 4.683572</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
        "      <td> 3.792342</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
        "      <td> 3.243505</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4</th>\n",
        "      <td> 1.616520</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "output_type": "pyout",
       "prompt_number": 5,
       "text": [
        "   activity\n",
        "0  3.655182\n",
        "1  4.683572\n",
        "2  3.792342\n",
        "3  3.243505\n",
        "4  1.616520"
       ]
      }
     ],
     "prompt_number": 5
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "fB"
     ],
     "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>activity</th>\n",
        "    </tr>\n",
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0 </th>\n",
        "      <td>  9.767125</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>1 </th>\n",
        "      <td> 11.808575</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2 </th>\n",
        "      <td>  6.911728</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3 </th>\n",
        "      <td>  3.653793</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>4 </th>\n",
        "      <td>  4.624176</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>5 </th>\n",
        "      <td>  4.086704</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>6 </th>\n",
        "      <td>  4.761538</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>7 </th>\n",
        "      <td> 10.870465</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>8 </th>\n",
        "      <td>  4.619190</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>9 </th>\n",
        "      <td>  7.627330</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>10</th>\n",
        "      <td>  7.699657</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>11</th>\n",
        "      <td> 12.236802</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>12</th>\n",
        "      <td>  7.178866</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>13</th>\n",
        "      <td> 12.890972</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>14</th>\n",
        "      <td> 11.637510</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>15</th>\n",
        "      <td> 10.865843</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
        "</div>"
       ],
       "output_type": "pyout",
       "prompt_number": 6,
       "text": [
        "     activity\n",
        "0    9.767125\n",
        "1   11.808575\n",
        "2    6.911728\n",
        "3    3.653793\n",
        "4    4.624176\n",
        "5    4.086704\n",
        "6    4.761538\n",
        "7   10.870465\n",
        "8    4.619190\n",
        "9    7.627330\n",
        "10   7.699657\n",
        "11  12.236802\n",
        "12   7.178866\n",
        "13  12.890972\n",
        "14  11.637510\n",
        "15  10.865843"
       ]
      }
     ],
     "prompt_number": 6
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "And what are their mean values."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "fA.mean()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 7,
       "text": [
        "activity    3.398224\n",
        "dtype: float64"
       ]
      }
     ],
     "prompt_number": 7
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "fB.mean()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "prompt_number": 8,
       "text": [
        "activity    8.202517\n",
        "dtype: float64"
       ]
      }
     ],
     "prompt_number": 8
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Is this difference significant? A simple way to explore this question is to use the bootstrap_plot function. I'll use this to subsample group B 5 elements at a time so that I can look at the distribution of the mean, median, and midpoints of this group."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "bs1 = figure(figsize=(10,7))\n",
      "bs1 = bootstrap_plot(fB, fig=bs1, size=5, samples=4368, color='grey')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAl0AAAGyCAYAAADeeHHhAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYIP/A5IKIyYk7VjMAAAAAElFTkSuQmCC\n"
      }
     ],
     "prompt_number": 9
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Hmm.... It looks like there is quite a bit of potential variation in the measured mean if you just choose a subset of 5 elements. I will repeat this operation explicity for the mean again and plot the mean of group A for comparison.\n",
      "\n",
      "I'll start by subsampling group B repetitively"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "sample_B = [mean(random.sample(B, 5)) for ii in range(4300)]"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 10
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Now I'll generate the histogram again but include the mean of group A on the plot:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "bs2 = hist(sample_B, range=(0,13.0), bins=30)\n",
      "ax = gca(); fig = gcf()\n",
      "ann = ax.annotate(\"group A mean\", xy=(fA.mean(), 0.0), xytext=(fA.mean()-3,500), \n",
      "            arrowprops=dict(arrowstyle=\"-|>\", connectionstyle=\"arc\", color='red'), color='red')\n",
      "t=title('distribution of the mean of group B')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAXgAAAEICAYAAABVv+9nAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nxYAnInJQDHgiIgfFgCciclAMeCIiB/X/ASc7/6d2Wo4AAAAASUVORK5CYII=\n"
      }
     ],
     "prompt_number": 11
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "meanCI = bootstrap.ci(data=B, statfunction=mean, alpha=0.05) # calculate 95% confidence intervals for the mean of group B"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 12
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# mark the 95% confidence intervals in green\n",
      "fig.axes[0].annotate(\"\", xy=(meanCI[0], 0.0), xytext=(meanCI[0],700), \n",
      "            arrowprops=dict(arrowstyle=\"->\", connectionstyle=\"arc\", color='green'), color='green')\n",
      "fig.axes[0].annotate(\"\", xy=(meanCI[1], 0.0), xytext=(meanCI[1],700), \n",
      "            arrowprops=dict(arrowstyle=\"->\", connectionstyle=\"arc\", color='green'), color='green')\n",
      "\n",
      "fig"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "pyout",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAXgAAAEICAYAAABVv+9nAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
       "prompt_number": 13,
       "text": [
        "<matplotlib.figure.Figure at 0x10899af50>"
       ]
      }
     ],
     "prompt_number": 13
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "I have calculated the confidence intervals on the mean value of group B and marked them on the histogram with green lines. It should now be clear whether the mean of group A is inside or outside of the 95% confidence interval of group B based upon the variation caused by different subsamples."
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "print \"The confidence interval for the group B mean if choosing 5 elements is:\", meanCI"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "output_type": "stream",
       "stream": "stdout",
       "text": [
        "The confidence interval for the group B mean if choosing 5 elements is: [ 6.69046781  9.73502992]\n"
       ]
      }
     ],
     "prompt_number": 14
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [],
     "language": "python",
     "metadata": {},
     "outputs": []
    }
   ],
   "metadata": {}
  }
 ]
}