Is there a way to make use of a JMP formula outside JMP that does not involve SAS or SAS Model Manager? If I do a linear regression in JMP with continuous variables and then want to make use of that model elsewhere, it's a simple step to put the linear equation into Excel or code it in to just about anything. However, for something more complicated like using nominal variables in regression or something incredibly complex like a boosted tree (in JMP Pro), making use of the prediction formula outside JMP by translating it into some other "language" is very tedious at best and approaching impossible at worst. I've seen "Get SAS Data Step for Formula Columns" and "Get MM SAS Data Step for Formula Columns", but this doesn't help if the model needs to be in Excel or any other system that performs calculations.
Unless I'm missing an important feature, JMP does not offer a way to "export" a formula to something common like VBA. I'm wondering if anyone knows about a third-party solution to interpret a JMP formula and translate input to output through the formula outside of JMP. Of course, if there is some way that doesn't involve a third-party solution, that would be even better.
I'm not hopeful that anything like this exists, but I'd be incredibly happy to find that it does. Maybe there is a business opportunity if no solution currently exists.
I also have exactly this problem and I'm still in need of a solution. It would be so helpful to have the ability to import complex formulas into Excel from the neural networks or boosted tree platforms (or all platforms for that matter). I tried translating a JMP formula into Excel and was not able to reproduce the results b/c of the model's complexity, my typos, etc. I Googled/researched solutions for a couple of hours and came up with nothing that worked so I had to result to using a simpler model which was not optimal for my problem. Any help would be greatly appreciated.
Thanks for the reply Dave. This is great to know. Sharing this at the Discovery Summit would be very nice. Otherwise, could you consult for us to produce a similar solution? thanks, Tee
Tee, this is probably not the appropriate forum to have that discussion - let's catch up on LinkedIn. Dave