cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
JMP is taking Discovery online, April 16 and 18. Register today and join us for interactive sessions featuring popular presentation topics, networking, and discussions with the experts.
Choose Language Hide Translation Bar
vincem
Level II

RE: how do i Fit a constrained longitudinal data analysis (cLDA) in JMP

Hi there

How do i fit a cLDA in JMP (via FIT command say) with the constraint that the group means are assumed to be equal at baseline (hence the “constrained” in cLDA). Hence if we had a model with Time and their interaction Group:Time as predictors theoretically  this model structure would impose that the group means are equal at baseline (i think?).

 

In cLDA baseline means are constrained to be equal between the randomised groups; a common baseline mean is assumed and fit across randomised groups. Here is the cDLA model fitted in R..hence i am aiming to implement the cDLA model in JMP via FIT command if possible.

 

Using the FIT data~ time+group.time doesn't seem to impose the common baseline mean constraints.

 

Any advice on how to fit such models in JMP will be appreciated.

 

regards Vince

2 REPLIES 2

RE: how do i Fit a constrained longitudinal data analysis (cLDA) in JMP

I am not familiar with this particular model but I believe that generally you should not include the group (categorical) term in the model to force a common intercept (baseline). You can include it in transformations such as cross terms to interaction effects. This form breaks the linear model hierarchy and JMP will complain but it will let you fit this model.

 

  1. Select Analyze > Fit Model.
  2. Select response column and click Y.
  3. Select the group and time columns, click Macros, and select Full Factorial.
  4. Select the group term in the Effects list and click Remove.
  5. Click Run.
  6. Click Continue when JMP raises issue about missing effect.
vincem
Level II

RE: how do i Fit a constrained longitudinal data analysis (cLDA) in JMP

Thanks Mark
Yes I did assume this to be the way to proceed but I don’t think this is correct . If you do a cross tab of group by time of the predicted values we should get equivalent baseline means - however I believe this isn’t the case.