How to correctly compute Odds Ratios from a Binomial Logit Generalized Linear Mixed Model (GLMM) in JMP Pro?

[#1]

I am fitting a Binomial Logit Generalized Linear Mixed Model (GLMM) in JMP Pro 18, with:

• binary outcome (1 = event, 0 = no event),

• several categorical predictors (factors),

• and a random intercept for subject ID to account for repeated measures / clustering.

The model converges and produces a table of Fixed Effects Parameter Estimates, which includes:

• Estimate (β)

• Standard Error

• Lower / Upper Confidence Limits (on the logit scale)

• p-values

However, JMP Pro does not provide Odds Ratios.

My questions:

1. In JMP Pro's GLMM (Binomial Logit), does “Estimate” in the Fixed Effects table represent the conditional log-odds ratio?

2. Is it correct and statistically valid to report OR = exp(Estimate) and CIs = exp(CI on the logit scale)?

3. If not, what is the recommended method for obtaining interpretable Odds Ratios from a GLMM in JMP?

4. Is there any official documentation or example demonstrating how ORs should be computed for GLMMs in JMP (or equivalently, in SAS PROC GLIMMIX)?

Any authoritative explanation or reference would be greatly appreciated.