for analysis not possible in JMP

[#1]

I feel like I have reached a stage where I have to do analysis which are not possible on JMP ( like doing a propensity matching, random effect for logistic regression, multiple imputation for categorical variables etc)

Even though I love JMP and will continue to use it .. I need to learn another statistical program to perform analysis which is not possible in JMP.

Since it requires quite a bit of time investment in learning a new program have been trying to explore options. Readers of this forum are probably  biased but I want to bring up for discussion.. Which would be a better program to learn for somebody who is fairly competent in JMP

SAS vs R

I have been searching online and most of the reviews are mixed although it seems like people are preferring R more than SAS. For a JMP user would there be any advantage in using SAS versus R?

[#2]

One advantage I can think of is the almost seamless integration of SAS and JMP in an organization's analytics workflow and ecosystem. Yes there is a JMP to R and back kind of integration as well...but it's no where near as elegant as the JMP to SAS and back bridge. After all, JMP is a SAS product so it's designed to play very nicely with SAS for just the types of situations you describe...where SAS has analytics capabilities not found natively in JMP products.

[#3]

@Sandeep123,  Have you had a chance to review this add-in Generalized Linear Mixed Model Add-in - JMP User Community?

From my perspective, JMP to R integration is very useful when I do work outside of JMP, but I do try to keep everything in the JSL file for documentation and reproducibility.  In the past I used SAS a bit, but found that R is more readily available in my field and so that governed my decision.  In my current role, I have access to SAS and find that, as Peter says, the JMP/SAS integration is very smooth.

[#4]

thank you. I did look at this add in. Haven't been able to work it the way I want so far...