I am using JMP 12.2 or JMP Pro 11. My problem is with the time it takes to run some models. Some of my data sets are 50,000 to 150,000 records and though I can do some basic logistic regression or ANOVA models, when I try to get the Odds Ratios, or run Cox Proportional Hazards models and generate Risk Ratios, it takes forever, if it even completes. Yesterday, I started one that I allowed to run for ~18 hrs and I finally shut JMP down and restarted so that I could do something different. Any thoughts on what I can do to speed this up?

Is there any way, once we have our final model results, to create the RR or OR estimates for only one or two variables in a multivariate model? For instance, I may have a model that has Group (50 levels), Month (12 levels), Age (11 levels), etc. My primary interest and reason for estimating the RR is to understand the impact of Age. When I am finished with the initial model, I have the p-values for each variable but is it possible to then generate the RR for only 1 variable simply to speed up the process?

Is there some other fix to this problem? I don’t think it a computer/processor speed/size issue but I am really frustrated…