<?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: how do I get positive predicted probabilities in GLIMMIX? in Statistical Procedures</title>
    <link>https://communities.sas.com/t5/Statistical-Procedures/how-do-I-get-positive-predicted-probabilities-in-GLIMMIX/m-p/453377#M23698</link>
    <description>&lt;P&gt;In the dataset&amp;nbsp;&lt;SPAN&gt;LBP_GLIMMIX_PRED3,&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&lt;SPAN&gt;&amp;nbsp;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&lt;SPAN&gt;&lt;EM&gt;pred_value&lt;/EM&gt; contains predicted values for each observation on the link (logit) scale, which ranges negative to positive.&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&lt;SPAN&gt;&lt;EM&gt;predprob&lt;/EM&gt; contains predicted values for each observation on the inverse link (proportion) scale, which ranges 0 to 1&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&lt;SPAN&gt;Are you looking at the correct variable? Does your &lt;EM&gt;predprob&lt;/EM&gt; contain negative values?&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
    <pubDate>Wed, 11 Apr 2018 21:31:02 GMT</pubDate>
    <dc:creator>sld</dc:creator>
    <dc:date>2018-04-11T21:31:02Z</dc:date>
    <item>
      <title>how do I get positive predicted probabilities in GLIMMIX?</title>
      <link>https://communities.sas.com/t5/Statistical-Procedures/how-do-I-get-positive-predicted-probabilities-in-GLIMMIX/m-p/453183#M23692</link>
      <description>&lt;P&gt;Hi all, I am estimating predicted rates and am using GLIMMIX to estimate the probability of receiving an image (0/1) based on several dummies.&amp;nbsp; I am using GLIMMIX to support the random effect of the clinician and address the clusters of patients nested in clinicians.&amp;nbsp;&amp;nbsp; The following syntax generates negative probabilities, which I suspect might be due to the distribution.&amp;nbsp; With LOGISTIC, I get probabilities btn&amp;nbsp; 0 and 1, but I can't control for the clustering or random effects of the clinician. (in SAS 9.4)&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;My principal question is, how can I rescale the probabilities in GLIMMIX to be btn 0 &amp;amp; 1?&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;Thank you!&lt;/P&gt;&lt;P&gt;&amp;nbsp;&lt;/P&gt;&lt;P&gt;PROC GLIMMIX DATA=out.low_back_final_Spec_Gen4 METHOD=LAPLACE NOCLPRINT;&lt;BR /&gt;CLASS&amp;nbsp; image_rfr_npi;&lt;BR /&gt;MODEL&amp;nbsp;&amp;nbsp; X_RAY_MRI_FLAG&amp;nbsp; = NPI_female Disable age AIDS ALCOHOL ANEMDEF ARTH BLDLOSS CHF CHRNLUNG COAG DEPRESS DM DMCX DRUG HTN HTNCX HYPOTHY LIVER LYMPH LYTES METS NEURO OBESE PARA PERIVASC PSYCH PULMCIRC RENLFAIL TUMOR ULCER VALVE WGHTLOSS Atlanta Bangor Boston Chattanooga Cleveland Dallas Des_Moines Elgin Greensboro Kingsport Madison Melrose_Park Milwaukee Morristown Nashville New_Brunswick New_Haven Omaha Orange_County Others Peoria Phoenix Pittsburgh Portland Raleigh San_Bernardino Seattle Springfield Sun_City Tampa Worcester&lt;/P&gt;&lt;P&gt;/&amp;nbsp; dist=binomial solution;&lt;BR /&gt;random intercept / type=vc subject=image_rfr_npi;&lt;BR /&gt;output out=LBP_GLIMMIX_PRED3 predicted=pred_value pred(ilink)=predprob;&lt;BR /&gt;COVTEST / WALD;&lt;BR /&gt;run;&lt;/P&gt;</description>
      <pubDate>Wed, 11 Apr 2018 12:58:08 GMT</pubDate>
      <guid>https://communities.sas.com/t5/Statistical-Procedures/how-do-I-get-positive-predicted-probabilities-in-GLIMMIX/m-p/453183#M23692</guid>
      <dc:creator>grateful</dc:creator>
      <dc:date>2018-04-11T12:58:08Z</dc:date>
    </item>
    <item>
      <title>Re: how do I get positive predicted probabilities in GLIMMIX?</title>
      <link>https://communities.sas.com/t5/Statistical-Procedures/how-do-I-get-positive-predicted-probabilities-in-GLIMMIX/m-p/453377#M23698</link>
      <description>&lt;P&gt;In the dataset&amp;nbsp;&lt;SPAN&gt;LBP_GLIMMIX_PRED3,&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&lt;SPAN&gt;&amp;nbsp;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&lt;SPAN&gt;&lt;EM&gt;pred_value&lt;/EM&gt; contains predicted values for each observation on the link (logit) scale, which ranges negative to positive.&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&lt;SPAN&gt;&lt;EM&gt;predprob&lt;/EM&gt; contains predicted values for each observation on the inverse link (proportion) scale, which ranges 0 to 1&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&lt;SPAN&gt;Are you looking at the correct variable? Does your &lt;EM&gt;predprob&lt;/EM&gt; contain negative values?&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Wed, 11 Apr 2018 21:31:02 GMT</pubDate>
      <guid>https://communities.sas.com/t5/Statistical-Procedures/how-do-I-get-positive-predicted-probabilities-in-GLIMMIX/m-p/453377#M23698</guid>
      <dc:creator>sld</dc:creator>
      <dc:date>2018-04-11T21:31:02Z</dc:date>
    </item>
  </channel>
</rss>

