<?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 Re: GEE logor=exch vs. type=exch in Statistical Procedures</title>
    <link>https://communities.sas.com/t5/Statistical-Procedures/GEE-logor-exch-vs-type-exch/m-p/94016#M4651</link>
    <description>&lt;HTML&gt;&lt;HEAD&gt;&lt;/HEAD&gt;&lt;BODY&gt;&lt;P&gt;This sounds like a job for Tech Support.&amp;nbsp; I have a guess, but it is probably wrong.&amp;nbsp; Any possibility that an OR is pathological for one of the waves, such that the log odds ratios are not exchangeable, but the correlations are?&amp;nbsp; By pathological, I mean a zero or some value greater than the machine arithmetic recognizes? This would result in errors when the alternating method is applied.&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Steve Denham&lt;/P&gt;&lt;/BODY&gt;&lt;/HTML&gt;</description>
    <pubDate>Mon, 15 Oct 2012 12:51:41 GMT</pubDate>
    <dc:creator>SteveDenham</dc:creator>
    <dc:date>2012-10-15T12:51:41Z</dc:date>
    <item>
      <title>GEE logor=exch vs. type=exch</title>
      <link>https://communities.sas.com/t5/Statistical-Procedures/GEE-logor-exch-vs-type-exch/m-p/94015#M4650</link>
      <description>&lt;HTML&gt;&lt;HEAD&gt;&lt;/HEAD&gt;&lt;BODY&gt;&lt;P&gt;I am trying to determine why I received the following errors when I ran GEE on my data using logor=exch and why I did not receive those errors when I ran my data with type=exch:&lt;/P&gt;&lt;P&gt;Error in log odds ratio regression computation.&lt;/P&gt;&lt;P&gt;Error in parameter estimate covariance computation.&lt;/P&gt;&lt;P&gt;Error in estimation routine.&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;The respective code is listed below:&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;proc genmod data=both;&lt;/P&gt;&lt;P&gt;class inhalyn (ref="0") grades (ref="3") wave uid / param=ref;&lt;/P&gt;&lt;P&gt;model inhalyn=grades / dist=bin link=logit;&lt;/P&gt;&lt;P&gt;repeated subject=uid / withinsubject=wave logor=exch;&lt;/P&gt;&lt;P&gt;estimate "1 vs 3" grades 1 0 / exp;&lt;/P&gt;&lt;P&gt;estimate "2 vs 3" grades 0 1 / exp;&lt;/P&gt;&lt;P&gt;run;&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;proc genmod data=both;&lt;/P&gt;&lt;P&gt;class inhalyn (ref="0") grades (ref="3") wave uid / param=ref;&lt;/P&gt;&lt;P&gt;model inhalyn=grades / dist=bin link=logit;&lt;/P&gt;&lt;P&gt;repeated subject=uid / withinsubject=wave type=exch;&lt;/P&gt;&lt;P&gt;estimate "1 vs 3" grades 1 0 / exp;&lt;/P&gt;&lt;P&gt;estimate "2 vs 3" grades 0 1 / exp;&lt;/P&gt;&lt;P&gt;run;&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;SAS was able to give me estimates and ORs when using type=exch. Inhalyn is a binary variable and grades is a 3-level categorical variable. I used the same code for other 3-level categorical variables and was able to get estimates and ORs when using logor=exch, so I'm trying to figure out what's different about this variable/scenario that would require type=exch instead of logor=exch.&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Your help is much appreciated!&lt;/P&gt;&lt;/BODY&gt;&lt;/HTML&gt;</description>
      <pubDate>Sat, 13 Oct 2012 00:35:17 GMT</pubDate>
      <guid>https://communities.sas.com/t5/Statistical-Procedures/GEE-logor-exch-vs-type-exch/m-p/94015#M4650</guid>
      <dc:creator>mcmag</dc:creator>
      <dc:date>2012-10-13T00:35:17Z</dc:date>
    </item>
    <item>
      <title>Re: GEE logor=exch vs. type=exch</title>
      <link>https://communities.sas.com/t5/Statistical-Procedures/GEE-logor-exch-vs-type-exch/m-p/94016#M4651</link>
      <description>&lt;HTML&gt;&lt;HEAD&gt;&lt;/HEAD&gt;&lt;BODY&gt;&lt;P&gt;This sounds like a job for Tech Support.&amp;nbsp; I have a guess, but it is probably wrong.&amp;nbsp; Any possibility that an OR is pathological for one of the waves, such that the log odds ratios are not exchangeable, but the correlations are?&amp;nbsp; By pathological, I mean a zero or some value greater than the machine arithmetic recognizes? This would result in errors when the alternating method is applied.&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Steve Denham&lt;/P&gt;&lt;/BODY&gt;&lt;/HTML&gt;</description>
      <pubDate>Mon, 15 Oct 2012 12:51:41 GMT</pubDate>
      <guid>https://communities.sas.com/t5/Statistical-Procedures/GEE-logor-exch-vs-type-exch/m-p/94016#M4651</guid>
      <dc:creator>SteveDenham</dc:creator>
      <dc:date>2012-10-15T12:51:41Z</dc:date>
    </item>
    <item>
      <title>Re: GEE logor=exch vs. type=exch</title>
      <link>https://communities.sas.com/t5/Statistical-Procedures/GEE-logor-exch-vs-type-exch/m-p/94017#M4652</link>
      <description>&lt;HTML&gt;&lt;HEAD&gt;&lt;/HEAD&gt;&lt;BODY&gt;&lt;P&gt;I think you're right!&lt;/P&gt;&lt;P&gt;&lt;/P&gt;&lt;P&gt;Thanks so much for your help.&lt;/P&gt;&lt;/BODY&gt;&lt;/HTML&gt;</description>
      <pubDate>Mon, 15 Oct 2012 19:49:18 GMT</pubDate>
      <guid>https://communities.sas.com/t5/Statistical-Procedures/GEE-logor-exch-vs-type-exch/m-p/94017#M4652</guid>
      <dc:creator>mcmag</dc:creator>
      <dc:date>2012-10-15T19:49:18Z</dc:date>
    </item>
  </channel>
</rss>

