<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic Brier score - Logistic Regression in Statistical Procedures</title>
    <link>https://communities.sas.com/t5/Statistical-Procedures/Brier-score-Logistic-Regression/m-p/207995#M11188</link>
    <description>&lt;HTML&gt;&lt;HEAD&gt;&lt;/HEAD&gt;&lt;BODY&gt;&lt;P&gt;Proc logistic returns brier score with fitstat option. My question - Can we set cutoff to check model calibration? What i understand it is sensitive to % of events. For example, for 50% event rate model, brier score less than 0.09 considers satisfactory. Does it exist for 10% event rate model? Does SAS generate scaled brier score?&lt;/P&gt;&lt;/BODY&gt;&lt;/HTML&gt;</description>
    <pubDate>Tue, 21 Jul 2015 21:48:58 GMT</pubDate>
    <dc:creator>Ujjawal</dc:creator>
    <dc:date>2015-07-21T21:48:58Z</dc:date>
    <item>
      <title>Brier score - Logistic Regression</title>
      <link>https://communities.sas.com/t5/Statistical-Procedures/Brier-score-Logistic-Regression/m-p/207995#M11188</link>
      <description>&lt;HTML&gt;&lt;HEAD&gt;&lt;/HEAD&gt;&lt;BODY&gt;&lt;P&gt;Proc logistic returns brier score with fitstat option. My question - Can we set cutoff to check model calibration? What i understand it is sensitive to % of events. For example, for 50% event rate model, brier score less than 0.09 considers satisfactory. Does it exist for 10% event rate model? Does SAS generate scaled brier score?&lt;/P&gt;&lt;/BODY&gt;&lt;/HTML&gt;</description>
      <pubDate>Tue, 21 Jul 2015 21:48:58 GMT</pubDate>
      <guid>https://communities.sas.com/t5/Statistical-Procedures/Brier-score-Logistic-Regression/m-p/207995#M11188</guid>
      <dc:creator>Ujjawal</dc:creator>
      <dc:date>2015-07-21T21:48:58Z</dc:date>
    </item>
    <item>
      <title>Re: Brier score - Logistic Regression</title>
      <link>https://communities.sas.com/t5/Statistical-Procedures/Brier-score-Logistic-Regression/m-p/207996#M11189</link>
      <description>&lt;HTML&gt;&lt;HEAD&gt;&lt;/HEAD&gt;&lt;BODY&gt;&lt;P&gt;&lt;SPAN style="color: #000000; font-family: arial, 'Arial Unicode MS', geneva, 'Lucida Grande', sans-serif; font-size: 13.4399995803833px; background-color: #ffffff;"&gt;In SAS website, it states that brier score &lt;/SPAN&gt;&lt;SPAN style="color: #000000; font-family: arial, 'Arial Unicode MS', geneva, 'Lucida Grande', sans-serif; font-size: 13.4399995803833px; background-color: #ffffff;"&gt;is the weighted squared difference between the predicted probabilities and their observed response levels.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN style="color: #000000; font-family: arial, 'Arial Unicode MS', geneva, 'Lucida Grande', sans-serif; font-size: 13.4399995803833px; background-color: #ffffff;"&gt;&lt;A class="active_link" href="http://support.sas.com/documentation/cdl/en/statug/63962/HTML/default/viewer.htm#statug_logistic_sect050.htm#statug.logistic.logdfitstat" title="http://support.sas.com/documentation/cdl/en/statug/63962/HTML/default/viewer.htm#statug_logistic_sect050.htm#statug.logistic.logdfitstat"&gt;SAS/STAT(R) 9.3 User's Guide&lt;/A&gt;&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN style="color: #000000; font-family: arial, 'Arial Unicode MS', geneva, 'Lucida Grande', sans-serif; font-size: 13.4399995803833px; background-color: #ffffff;"&gt;&lt;BR /&gt;&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN style="color: #000000; font-family: arial, 'Arial Unicode MS', geneva, 'Lucida Grande', sans-serif; font-size: 13.4399995803833px; background-color: #ffffff;"&gt;Does it take into account the event rate (% of events)? Can i safely use cutoff .09? Brier score less than .09 is acceptable.&lt;BR /&gt;&lt;/SPAN&gt;&lt;/P&gt;&lt;/BODY&gt;&lt;/HTML&gt;</description>
      <pubDate>Wed, 22 Jul 2015 11:25:08 GMT</pubDate>
      <guid>https://communities.sas.com/t5/Statistical-Procedures/Brier-score-Logistic-Regression/m-p/207996#M11189</guid>
      <dc:creator>Ujjawal</dc:creator>
      <dc:date>2015-07-22T11:25:08Z</dc:date>
    </item>
  </channel>
</rss>

