Subscribe Bookmark RSS Feed



Aug 19, 2014

Random Coefficient Models: How to Model Longitudinal and Hierarchical Data in JMP® Pro

Elizabeth Claassen, PhD, JMP Research Statistician Tester, SAS

Chris Gotwalt, PhD, JMP Director of Statistical Research and Development, SAS

The addition of the Mixed Model personality of Fit Model (Fit Mixed) has greatly expanded the ability of JMP Pro to analyse longitudinal data that consists of repeated measurements taken over time. It is perhaps not yet fully appreciated how useful the platform is across a wide variety of subject matter areas, including the social sciences, product reliability and agriculture. The same underlying methodology in the platform can be used to model individual growth/degradation curves or to model experimental units that are organised in a hierarchy, such as students within schools or individual plants within plots. Because the methods for modelling effects over time developed separately in several different disciplines, there can be big differences in terminology for what are basically the same models. For example, random coefficient models, hierarchical linear models (HLMs) and hierarchical Bayes models are essentially the same models, and all can be fit using Fit Mixed in JMP Pro 11 and JMP Pro 12. In this presentation we demonstrate the capabilities of Fit Mixed for random coefficient models using examples from the pharmaceutical industry and social sciences. This includes making the “translation” from SAS PROC MIXED for JMP users accustomed to fitting mixed models in SAS.


Thanks for cross walking the JMP Mixed Model Platform to SAS PROC Mixed. The JMP Platform for random intercepts and more generally, mixed models is good.  I would like to see the JMP platform capable of estimating coefficients for categorical outcomes in the future as well. Thanks again!  


Thank you for your kind words.

I want to clarify your request. Are you wanting random coefficients for categorical factors? Or are you looking for mixed models with binomial-type data, so logistic regression with random effects?



Hi Elizabeth,

I wish I had been able to attend your presentation last week but I think I had to be somewhere else.

I am pretty sure that @mpg is asking for logistic regression with random effects. Maybe that could then be extended to ordinal logistic regression as well.




Thanks, Phil. That's what I thought, but I wanted to be sure. GLMMs and GEEs (mixed models with non-normal response distributions) have been requested before. I'll add another vote for them.




It gets my vote! I had to do some coursework recently for a Mixed Models module of my Applied Stats Masters and it would have been good to have been able to stay in JMP for the logistic regression part. Instead I had to waste my time working out the bugs in some dodgy R code from the lecture notes!


Hello Elizabeth and Phil,

I apologize for delayed response, have been moving around.

Yes I am requesting that future releases of JMP consider adding GLMM's to the mixed modeling platform.  On the other hand, I was thinking when I have a bit of free time I would make an effort to write a JMP add-in that uses R to estimate mixed models with non-normal responses.

Any thoughts on creating that add-in?


Matt Goodlaw



I've entered your "vote" for GLMMs into our tracking. I can't promise anything, but that request does seem to be rising to the top of the list for inclusion in a near-er future release.

I know others have considered taking Russ Wolfinger's original GLIMMIX macro, which wrapped around PROC MIXED to do GLMMs and turn it into an Add-In by wrapping around Fit Model-Mixed Model in JMP. I don't know if it has never happened because of time or difficulty.

Thanks for your interest!



Thanks, Elizabeth.  Unfortunately, I no longer have SAS Base available to me....Though I would be interested in that add-in if it comes together.  In the meantime, if you are interested I will let you know my progress with the R option...



Article Labels
Article Tags