fnielsen/Nielsen2014Data_femalehurricanes.ipynb

## Nielsen2014Data_femalehurricanes.ipynb
{
 "metadata": {
  "name": "",
  "signature": "sha256:5a2a3d3914df716daf03cc8e36e5f3e371e5e53ac92bc4ba14a572fbf694f60c"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Female hurricanes are deadlier than male hurricanes\n",
      "===============================================\n",
      "A recent PNAS study has gained media attention. The original study is available from:\n",
      "\n",
      "http://www.pnas.org/content/early/2014/05/29/1402786111\n",
      "\n",
      "...but unfortunately paywalled.\n",
      "\n",
      "The conclusions drawn from the study seems to be that people do not take hurricanes seriously if they are given female names.\n",
      "\n",
      "The data is available as supplementary material in the form of an Excel sheet, that can be loaded with Python Pandas library (remember to skip the footer). Below is a reanalysis of the data. \n",
      "\n",
      "/Finn \u00c5rup Nielsen, GPLv3, CC-BY-SA"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "import matplotlib.pyplot as plt\n",
      "import numpy as np\n",
      "import pandas as pd\n",
      "import statsmodels.formula.api as smf\n",
      "import urllib\n",
      "\n",
      "%matplotlib inline"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 53
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Download the data from the PNAS site\n",
      "url = 'http://www.pnas.org/content/suppl/2014/05/30/1402786111.DCSupplemental/pnas.1402786111.sd01.xlsx'\n",
      "filename = 'pnas.1402786111.sd01.xlsx'\n",
      "urllib.urlretrieve(url, filename=filename)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 4,
       "text": [
        "('pnas.1402786111.sd01.xlsx', <httplib.HTTPMessage instance at 0x29e58c0>)"
       ]
      }
     ],
     "prompt_number": 4
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "# Read the data from the Excel file\n",
      "data = pd.read_excel(filename, 'Archival Study', skipfooter=6)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 43
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "def scatter(data, x='MasFem', y='alldeaths'):\n",
      "    \"\"\"\"plot Simple scatter of femininity and deaths.\"\"\"\n",
      "    plt.scatter(data[x], data[y], marker='.')\n",
      "    if x == 'MasFem' and y == 'alldeaths':\n",
      "        plt.title('Hurricane deaths versus Femininity')\n",
      "        plt.xlabel('Femininity')\n",
      "        plt.ylabel('Deaths')"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [],
     "prompt_number": 24
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "scatter(data)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAYYAAAEXCAYAAACpuuMDAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\npMBgICIiBQYDEREpMBiIiEjh/wPcxt3Afo6mhgAAAABJRU5ErkJggg==\n",
       "text": [
        "<matplotlib.figure.Figure at 0x5cc4ed0>"
       ]
      }
     ],
     "prompt_number": 44
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "The interpretation of the above plot based on the eye is that the gender effect (if any) is driven by a few female hurricanes. "
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Male hurricane names was introduced in 1979 according to \n",
      "http://www.washingtonpost.com/blogs/capital-weather-gang/wp/2014/06/02/female-named-hurricanes-kill-more-than-male-because-people-dont-respect-them-study-finds/\n",
      "We can exclude data before 1979:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "data_since1979 = data.ix[data['Year']>=1979, :]\n",
      "scatter(data_since1979)"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "metadata": {},
       "output_type": "display_data",
       "png": "iVBORw0KGgoAAAANSUhEUgAAAYYAAAEXCAYAAACpuuMDAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nSOX/AToF4G79rYETAAAAAElFTkSuQmCC\n",
       "text": [
        "<matplotlib.figure.Figure at 0x5e191d0>"
       ]
      }
     ],
     "prompt_number": 55
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "This plot now shows less of a gender effect (if any). Indeed there seems only to be a single female hurricane (Sandy of 2012) to drag the data. \n",
      "\n",
      "Statistical analyses\n",
      "------------------\n",
      "We can try to do a statistical analysis, first with an ordinary linear model with the category as a co-variate:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "smf.ols(formula='alldeaths ~ Gender_MF + Category', data=data_since1979).fit().summary()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<table class=\"simpletable\">\n",
        "<caption>OLS Regression Results</caption>\n",
        "<tr>\n",
        "  <th>Dep. Variable:</th>        <td>alldeaths</td>    <th>  R-squared:         </th> <td>   0.056</td>\n",
        "</tr>\n",
        "<tr>\n",
        "  <th>Model:</th>                   <td>OLS</td>       <th>  Adj. R-squared:    </th> <td>   0.019</td>\n",
        "</tr>\n",
        "<tr>\n",
        "  <th>Method:</th>             <td>Least Squares</td>  <th>  F-statistic:       </th> <td>   1.525</td>\n",
        "</tr>\n",
        "<tr>\n",
        "  <th>Date:</th>             <td>Tue, 03 Jun 2014</td> <th>  Prob (F-statistic):</th>  <td> 0.227</td> \n",
        "</tr>\n",
        "<tr>\n",
        "  <th>Time:</th>                 <td>17:56:14</td>     <th>  Log-Likelihood:    </th> <td> -253.53</td>\n",
        "</tr>\n",
        "<tr>\n",
        "  <th>No. Observations:</th>      <td>    54</td>      <th>  AIC:               </th> <td>   513.1</td>\n",
        "</tr>\n",
        "<tr>\n",
        "  <th>Df Residuals:</th>          <td>    51</td>      <th>  BIC:               </th> <td>   519.0</td>\n",
        "</tr>\n",
        "<tr>\n",
        "  <th>Df Model:</th>              <td>     2</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>Intercept</th> <td>    2.6907</td> <td>    8.972</td> <td>    0.300</td> <td> 0.765</td> <td>  -15.322    20.703</td>\n",
        "</tr>\n",
        "<tr>\n",
        "  <th>Gender_MF</th> <td>    1.7037</td> <td>    7.414</td> <td>    0.230</td> <td> 0.819</td> <td>  -13.180    16.587</td>\n",
        "</tr>\n",
        "<tr>\n",
        "  <th>Category</th>  <td>    6.4217</td> <td>    3.709</td> <td>    1.731</td> <td> 0.089</td> <td>   -1.025    13.869</td>\n",
        "</tr>\n",
        "</table>\n",
        "<table class=\"simpletable\">\n",
        "<tr>\n",
        "  <th>Omnibus:</th>       <td>66.017</td> <th>  Durbin-Watson:     </th> <td>   1.281</td> \n",
        "</tr>\n",
        "<tr>\n",
        "  <th>Prob(Omnibus):</th> <td> 0.000</td> <th>  Jarque-Bera (JB):  </th> <td> 507.894</td> \n",
        "</tr>\n",
        "<tr>\n",
        "  <th>Skew:</th>          <td> 3.331</td> <th>  Prob(JB):          </th> <td>5.15e-111</td>\n",
        "</tr>\n",
        "<tr>\n",
        "  <th>Kurtosis:</th>      <td>16.467</td> <th>  Cond. No.          </th> <td>    6.68</td> \n",
        "</tr>\n",
        "</table>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 52,
       "text": [
        "<class 'statsmodels.iolib.summary.Summary'>\n",
        "\"\"\"\n",
        "                            OLS Regression Results                            \n",
        "==============================================================================\n",
        "Dep. Variable:              alldeaths   R-squared:                       0.056\n",
        "Model:                            OLS   Adj. R-squared:                  0.019\n",
        "Method:                 Least Squares   F-statistic:                     1.525\n",
        "Date:                Tue, 03 Jun 2014   Prob (F-statistic):              0.227\n",
        "Time:                        17:56:14   Log-Likelihood:                -253.53\n",
        "No. Observations:                  54   AIC:                             513.1\n",
        "Df Residuals:                      51   BIC:                             519.0\n",
        "Df Model:                           2                                         \n",
        "==============================================================================\n",
        "                 coef    std err          t      P>|t|      [95.0% Conf. Int.]\n",
        "------------------------------------------------------------------------------\n",
        "Intercept      2.6907      8.972      0.300      0.765       -15.322    20.703\n",
        "Gender_MF      1.7037      7.414      0.230      0.819       -13.180    16.587\n",
        "Category       6.4217      3.709      1.731      0.089        -1.025    13.869\n",
        "==============================================================================\n",
        "Omnibus:                       66.017   Durbin-Watson:                   1.281\n",
        "Prob(Omnibus):                  0.000   Jarque-Bera (JB):              507.894\n",
        "Skew:                           3.331   Prob(JB):                    5.15e-111\n",
        "Kurtosis:                      16.467   Cond. No.                         6.68\n",
        "==============================================================================\n",
        "\"\"\""
       ]
      }
     ],
     "prompt_number": 52
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "The above results yielded no particular noticeable result for the gender effect. We can try to transform the response variable with a log+1 and see what happens:"
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [
      "smf.ols(formula='np.log(alldeaths+1) ~ Gender_MF + Category', data=data_since1979).fit().summary()"
     ],
     "language": "python",
     "metadata": {},
     "outputs": [
      {
       "html": [
        "<table class=\"simpletable\">\n",
        "<caption>OLS Regression Results</caption>\n",
        "<tr>\n",
        "  <th>Dep. Variable:</th>    <td>np.log(alldeaths + 1)</td> <th>  R-squared:         </th> <td>   0.165</td>\n",
        "</tr>\n",
        "<tr>\n",
        "  <th>Model:</th>                     <td>OLS</td>          <th>  Adj. R-squared:    </th> <td>   0.132</td>\n",
        "</tr>\n",
        "<tr>\n",
        "  <th>Method:</th>               <td>Least Squares</td>     <th>  F-statistic:       </th> <td>   5.031</td>\n",
        "</tr>\n",
        "<tr>\n",
        "  <th>Date:</th>               <td>Tue, 03 Jun 2014</td>    <th>  Prob (F-statistic):</th>  <td>0.0101</td> \n",
        "</tr>\n",
        "<tr>\n",
        "  <th>Time:</th>                   <td>18:22:29</td>        <th>  Log-Likelihood:    </th> <td> -82.435</td>\n",
        "</tr>\n",
        "<tr>\n",
        "  <th>No. Observations:</th>        <td>    54</td>         <th>  AIC:               </th> <td>   170.9</td>\n",
        "</tr>\n",
        "<tr>\n",
        "  <th>Df Residuals:</th>            <td>    51</td>         <th>  BIC:               </th> <td>   176.8</td>\n",
        "</tr>\n",
        "<tr>\n",
        "  <th>Df Model:</th>                <td>     2</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>Intercept</th> <td>    1.0178</td> <td>    0.377</td> <td>    2.696</td> <td> 0.009</td> <td>    0.260     1.776</td>\n",
        "</tr>\n",
        "<tr>\n",
        "  <th>Gender_MF</th> <td>    0.1150</td> <td>    0.312</td> <td>    0.369</td> <td> 0.714</td> <td>   -0.511     0.741</td>\n",
        "</tr>\n",
        "<tr>\n",
        "  <th>Category</th>  <td>    0.4916</td> <td>    0.156</td> <td>    3.150</td> <td> 0.003</td> <td>    0.178     0.805</td>\n",
        "</tr>\n",
        "</table>\n",
        "<table class=\"simpletable\">\n",
        "<tr>\n",
        "  <th>Omnibus:</th>       <td> 3.562</td> <th>  Durbin-Watson:     </th> <td>   1.621</td>\n",
        "</tr>\n",
        "<tr>\n",
        "  <th>Prob(Omnibus):</th> <td> 0.168</td> <th>  Jarque-Bera (JB):  </th> <td>   3.024</td>\n",
        "</tr>\n",
        "<tr>\n",
        "  <th>Skew:</th>          <td> 0.579</td> <th>  Prob(JB):          </th> <td>   0.221</td>\n",
        "</tr>\n",
        "<tr>\n",
        "  <th>Kurtosis:</th>      <td> 3.033</td> <th>  Cond. No.          </th> <td>    6.68</td>\n",
        "</tr>\n",
        "</table>"
       ],
       "metadata": {},
       "output_type": "pyout",
       "prompt_number": 62,
       "text": [
        "<class 'statsmodels.iolib.summary.Summary'>\n",
        "\"\"\"\n",
        "                              OLS Regression Results                             \n",
        "=================================================================================\n",
        "Dep. Variable:     np.log(alldeaths + 1)   R-squared:                       0.165\n",
        "Model:                               OLS   Adj. R-squared:                  0.132\n",
        "Method:                    Least Squares   F-statistic:                     5.031\n",
        "Date:                   Tue, 03 Jun 2014   Prob (F-statistic):             0.0101\n",
        "Time:                           18:22:29   Log-Likelihood:                -82.435\n",
        "No. Observations:                     54   AIC:                             170.9\n",
        "Df Residuals:                         51   BIC:                             176.8\n",
        "Df Model:                              2                                         \n",
        "==============================================================================\n",
        "                 coef    std err          t      P>|t|      [95.0% Conf. Int.]\n",
        "------------------------------------------------------------------------------\n",
        "Intercept      1.0178      0.377      2.696      0.009         0.260     1.776\n",
        "Gender_MF      0.1150      0.312      0.369      0.714        -0.511     0.741\n",
        "Category       0.4916      0.156      3.150      0.003         0.178     0.805\n",
        "==============================================================================\n",
        "Omnibus:                        3.562   Durbin-Watson:                   1.621\n",
        "Prob(Omnibus):                  0.168   Jarque-Bera (JB):                3.024\n",
        "Skew:                           0.579   Prob(JB):                        0.221\n",
        "Kurtosis:                       3.033   Cond. No.                         6.68\n",
        "==============================================================================\n",
        "\"\"\""
       ]
      }
     ],
     "prompt_number": 62
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "This transform yielded neither an effect."
     ]
    },
    {
     "cell_type": "markdown",
     "metadata": {},
     "source": [
      "Conclusion\n",
      "---------\n",
      "So are we to conclude that the effect is questionable? The researchers left out Katrina and Audrey. I wonder what the results would be with these storms included. "
     ]
    },
    {
     "cell_type": "code",
     "collapsed": false,
     "input": [],
     "language": "python",
     "metadata": {},
     "outputs": []
    }
   ],
   "metadata": {}
  }
 ]
}