<?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: Output from Glimmix procedure, how to interpretate between estimate of &amp;quot;&amp;gt;999.999&amp;quot; v in Statistical Procedures</title>
    <link>https://communities.sas.com/t5/Statistical-Procedures/Output-from-Glimmix-procedure-how-to-interpretate-between/m-p/940375#M46941</link>
    <description>&lt;P&gt;This might be due to the quasi-complete separation.&lt;/P&gt;
&lt;P&gt;I see that you opened a case with SAS Technical Support. I will provide more information in the ticket if needed.&lt;/P&gt;
&lt;P&gt;Thanks,&lt;/P&gt;
&lt;P&gt;Jill&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
    <pubDate>Wed, 21 Aug 2024 18:52:59 GMT</pubDate>
    <dc:creator>jiltao</dc:creator>
    <dc:date>2024-08-21T18:52:59Z</dc:date>
    <item>
      <title>Output from Glimmix procedure, how to interpretate between estimate of "&gt;999.999" vs  "."</title>
      <link>https://communities.sas.com/t5/Statistical-Procedures/Output-from-Glimmix-procedure-how-to-interpretate-between/m-p/940362#M46938</link>
      <description>&lt;P&gt;Hi,&lt;/P&gt;&lt;P&gt;I ran the multinominal regression model with GLIMMIX procedure. The outcome variable has 7 categories and some of the categories contain small frequency counts or no event at all. The data is very sparse. On the Odds Ratio estimates output, there are estimates labelled as either "&amp;lt;0.001" or "&amp;gt;999.999". On the 95% confidence limits, some of the estimates are printed as "."&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;What are the differences between "&amp;gt;999.999" or "&amp;lt;0.0001" vs "."&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;In the case where I obtain OR='&amp;lt;0.001', the beta estimate is a very large negative number (example -9.6)&lt;/P&gt;&lt;P&gt;In the case where I obtain OR='&amp;gt;999.999", the beta estimate is a very large positive number (example, 12.87)&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;I believe "&amp;gt;999.999' and "&amp;lt;0.001" are SAS formats to label extremely large or small values, respectively . What about in place where I got "." Does "." means that the value is not estimable or exp(beta)=infinity?&amp;nbsp;&lt;/P&gt;&lt;P&gt;Is there a citation or documentation on these?&amp;nbsp;&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;Thank you&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;/P&gt;&lt;P&gt;Thank you so much!&lt;/P&gt;</description>
      <pubDate>Wed, 21 Aug 2024 18:08:41 GMT</pubDate>
      <guid>https://communities.sas.com/t5/Statistical-Procedures/Output-from-Glimmix-procedure-how-to-interpretate-between/m-p/940362#M46938</guid>
      <dc:creator>chouj</dc:creator>
      <dc:date>2024-08-21T18:08:41Z</dc:date>
    </item>
    <item>
      <title>Re: Output from Glimmix procedure, how to interpretate between estimate of "&gt;999.999" v</title>
      <link>https://communities.sas.com/t5/Statistical-Procedures/Output-from-Glimmix-procedure-how-to-interpretate-between/m-p/940375#M46941</link>
      <description>&lt;P&gt;This might be due to the quasi-complete separation.&lt;/P&gt;
&lt;P&gt;I see that you opened a case with SAS Technical Support. I will provide more information in the ticket if needed.&lt;/P&gt;
&lt;P&gt;Thanks,&lt;/P&gt;
&lt;P&gt;Jill&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Wed, 21 Aug 2024 18:52:59 GMT</pubDate>
      <guid>https://communities.sas.com/t5/Statistical-Procedures/Output-from-Glimmix-procedure-how-to-interpretate-between/m-p/940375#M46941</guid>
      <dc:creator>jiltao</dc:creator>
      <dc:date>2024-08-21T18:52:59Z</dc:date>
    </item>
    <item>
      <title>Re: Output from Glimmix procedure, how to interpretate between estimate of "&gt;999.999" v</title>
      <link>https://communities.sas.com/t5/Statistical-Procedures/Output-from-Glimmix-procedure-how-to-interpretate-between/m-p/940395#M46946</link>
      <description>&lt;P&gt;For some details you can try looking at the TEMPLATE definitions used to create the output.&lt;/P&gt;
&lt;P&gt;For example the&amp;nbsp; Oddsratio template for Glimmix has elements such as&lt;/P&gt;
&lt;PRE&gt;define Estimate;                                                        
         header = "Estimate";                                                 
         translate _val_=. into " Non-est";                                   
         format = oddsr8.3;                                                   
      end; 
&lt;/PRE&gt;
&lt;P&gt;And you can look up the Oddsrw. format to find:&lt;/P&gt;
&lt;BLOCKQUOTE&gt;
&lt;DIV id="n1trofl64hi1m9n0zouv7x7xwbii" class="xisDoc-comparisons"&gt;
&lt;P class="xisDoc-paragraph"&gt;The ODDSR&lt;EM class="xisDoc-userSuppliedValue"&gt;w&lt;/EM&gt;.&lt;EM class="xisDoc-userSuppliedValue"&gt;d&lt;/EM&gt; format follows the rules for the &lt;EM class="xisDoc-userSuppliedValue"&gt;w&lt;/EM&gt;.&lt;EM class="xisDoc-userSuppliedValue"&gt;d&lt;/EM&gt; format, except in these conditions:&lt;/P&gt;
&lt;UL class="xisDoc-listUnordered"&gt;
&lt;LI class="xisDoc-item"&gt;values from –1e-12 to 10&lt;SUP class="xisDoc-superscript"&gt;-d&lt;/SUP&gt; –eps are displayed as less than “0.0...01”&lt;/LI&gt;
&lt;LI class="xisDoc-item"&gt;values that are greater than or equal to 10&lt;SUP class="xisDoc-superscript"&gt;w-d-2&lt;/SUP&gt; are displayed as greater than “999.999”&lt;/LI&gt;
&lt;/UL&gt;
&lt;/DIV&gt;
&lt;/BLOCKQUOTE&gt;
&lt;P&gt;Special values are shown in the documentation but .&amp;nbsp; is still . (missing). So is suspect something was not able to be calculated for your data and options chosen. Ratios for example require two values. If one is missing no ratio can be calculated.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Wed, 21 Aug 2024 20:35:21 GMT</pubDate>
      <guid>https://communities.sas.com/t5/Statistical-Procedures/Output-from-Glimmix-procedure-how-to-interpretate-between/m-p/940395#M46946</guid>
      <dc:creator>ballardw</dc:creator>
      <dc:date>2024-08-21T20:35:21Z</dc:date>
    </item>
    <item>
      <title>Re: Output from Glimmix procedure, how to interpretate between estimate of "&gt;999.999" v</title>
      <link>https://communities.sas.com/t5/Statistical-Procedures/Output-from-Glimmix-procedure-how-to-interpretate-between/m-p/940432#M46956</link>
      <description>Maybe that is a value scale problem.&lt;BR /&gt;Make variable have more scale unit:&lt;BR /&gt;x=x*100;&lt;BR /&gt;x=x*1000;</description>
      <pubDate>Thu, 22 Aug 2024 02:51:19 GMT</pubDate>
      <guid>https://communities.sas.com/t5/Statistical-Procedures/Output-from-Glimmix-procedure-how-to-interpretate-between/m-p/940432#M46956</guid>
      <dc:creator>Ksharp</dc:creator>
      <dc:date>2024-08-22T02:51:19Z</dc:date>
    </item>
    <item>
      <title>Re: Output from Glimmix procedure, how to interpretate between estimate of "&gt;999.999" v</title>
      <link>https://communities.sas.com/t5/Statistical-Procedures/Output-from-Glimmix-procedure-how-to-interpretate-between/m-p/941426#M47000</link>
      <description>&lt;P&gt;Thank you so much! This is very helpful&lt;/P&gt;</description>
      <pubDate>Wed, 28 Aug 2024 13:52:20 GMT</pubDate>
      <guid>https://communities.sas.com/t5/Statistical-Procedures/Output-from-Glimmix-procedure-how-to-interpretate-between/m-p/941426#M47000</guid>
      <dc:creator>chouj</dc:creator>
      <dc:date>2024-08-28T13:52:20Z</dc:date>
    </item>
  </channel>
</rss>

