I have several dependent (let's say Y1, Y2, Y3....Y30) and several hundred predictors (e.g. x1....x600).
Using stepwise regression, I could find multiple regression models for each Ys but we want to narrow down number of meaningful x variables into down to 10.
Is there a way to find these 10 best x variables that would best represent Y1 to Y30 at same time using JMP? I think this type of analysis of optimization of multivariate model but if I am wrong please correct me.
I am more than happy to buy books and study the subject, so please let me know what this is called.
I do not know if it is available in JMP, but in SAS there is a procedure (PROC PLS) which is specifically designed to deal with this sort of situation. Google the documentation to see if that procedure (predictive partial least squares) would be useful in your situation. Then the folks in this forum that know JMP way better than I do might be able to point you at the right steps to implement this.