topic GLIMMIX for repeated measures and multinomial (ordered) response in Statistical Procedures https://communities.sas.com/t5/Statistical-Procedures/GLIMMIX-for-repeated-measures-and-multinomial-ordered-response/m-p/207497#M11153 <HTML><HEAD></HEAD><BODY><P>Looking for some interpretation help with a GLIMMIX output (as most the documentation is from logistic models). </P><P></P><P>The very basic bivariable model is as follows: </P><P>proc glimmix data=have noclprint noitprint method=quad gradient ; </P><P>&nbsp; class study_id ; </P><P>&nbsp; model score=race / solution dist=mult link=clogit ; </P><P>&nbsp; random intercept / type=un subject=study_id ; </P><P>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; covtest 'Need Random Intercept?' 0 ; </P><P>&nbsp; where race ^in(.,88,99) ; </P><P>run ; </P><P></P><P>I have 5500 observations, with a total of 1795 individuals (study_id) observed over time. Participants have multiple scores (range: 0-4) over time. I have made no assumption about the distance between adjacent levels, but they are ordered (i.e. 0=not present, 4=very bad levels present). Race is ethnic race (i.e. caucasian, African-American etc) with 5 levels. I am modeling the probabilities of levels of score having lower ordered values in the response profile table. </P><P></P><P>Once run, I have 4 intercepts&nbsp; (0-3), and 5 estimates for race. Numbers below are just made up. </P><P></P><P>Effect&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Score&nbsp;&nbsp;&nbsp;&nbsp; Race&nbsp;&nbsp;&nbsp;&nbsp; Estimate </P><P>Intercept&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0.5</P><P>Intercept&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 1&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0.6</P><P>Intercept&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 2&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0.7</P><P>Intercept&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 3&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0.8</P><P>Race&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Race1&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0.09</P><P>Race&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Race 2&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0.1</P><P>Race&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Race3&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0.11</P><P>Race&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Race4&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0.12</P><P>Race&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Race5&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0.13</P><P></P><P>I would like to be able to interpret this (eventually) in a multivariable model as well, then present it in a logical and meaningful way. Any direction would be appreciated.&nbsp;&nbsp;&nbsp;&nbsp; </P></BODY></HTML> Tue, 21 Jul 2015 16:41:41 GMT toneill 2015-07-21T16:41:41Z GLIMMIX for repeated measures and multinomial (ordered) response https://communities.sas.com/t5/Statistical-Procedures/GLIMMIX-for-repeated-measures-and-multinomial-ordered-response/m-p/207497#M11153 <HTML><HEAD></HEAD><BODY><P>Looking for some interpretation help with a GLIMMIX output (as most the documentation is from logistic models). </P><P></P><P>The very basic bivariable model is as follows: </P><P>proc glimmix data=have noclprint noitprint method=quad gradient ; </P><P>&nbsp; class study_id ; </P><P>&nbsp; model score=race / solution dist=mult link=clogit ; </P><P>&nbsp; random intercept / type=un subject=study_id ; </P><P>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; covtest 'Need Random Intercept?' 0 ; </P><P>&nbsp; where race ^in(.,88,99) ; </P><P>run ; </P><P></P><P>I have 5500 observations, with a total of 1795 individuals (study_id) observed over time. Participants have multiple scores (range: 0-4) over time. I have made no assumption about the distance between adjacent levels, but they are ordered (i.e. 0=not present, 4=very bad levels present). Race is ethnic race (i.e. caucasian, African-American etc) with 5 levels. I am modeling the probabilities of levels of score having lower ordered values in the response profile table. </P><P></P><P>Once run, I have 4 intercepts&nbsp; (0-3), and 5 estimates for race. Numbers below are just made up. </P><P></P><P>Effect&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Score&nbsp;&nbsp;&nbsp;&nbsp; Race&nbsp;&nbsp;&nbsp;&nbsp; Estimate </P><P>Intercept&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0.5</P><P>Intercept&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 1&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0.6</P><P>Intercept&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 2&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0.7</P><P>Intercept&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 3&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0.8</P><P>Race&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Race1&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0.09</P><P>Race&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Race 2&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0.1</P><P>Race&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Race3&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0.11</P><P>Race&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Race4&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0.12</P><P>Race&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Race5&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 0.13</P><P></P><P>I would like to be able to interpret this (eventually) in a multivariable model as well, then present it in a logical and meaningful way. Any direction would be appreciated.&nbsp;&nbsp;&nbsp;&nbsp; </P></BODY></HTML> Tue, 21 Jul 2015 16:41:41 GMT https://communities.sas.com/t5/Statistical-Procedures/GLIMMIX-for-repeated-measures-and-multinomial-ordered-response/m-p/207497#M11153 toneill 2015-07-21T16:41:41Z Re: GLIMMIX for repeated measures and multinomial (ordered) response https://communities.sas.com/t5/Statistical-Procedures/GLIMMIX-for-repeated-measures-and-multinomial-ordered-response/m-p/207498#M11154 <HTML><HEAD></HEAD><BODY><P>The key will be to translate the estimates into odds ratios that describe what is going on, and that isn't always easy.&nbsp; This is a case where ESTIMATE statements are critical to calculating customized odds ratios.&nbsp; Do a search of this site on 'glimmix multinomial estimate' as I am sure there are answers out there by people who are better at this than I am.</P><P></P><P>Steve Denham</P></BODY></HTML> Wed, 22 Jul 2015 12:26:17 GMT https://communities.sas.com/t5/Statistical-Procedures/GLIMMIX-for-repeated-measures-and-multinomial-ordered-response/m-p/207498#M11154 SteveDenham 2015-07-22T12:26:17Z