topic Re: Margin of error of predictor screening in Discussions

Margin of error of predictor screening

melina_d — Thu, 08 Jun 2023 16:29:06 GMT

Hi!

I was wondering if I can get the margin of error of a predictor screening model I run to be able to compare the portion of the predictors across different groups within one sample -by running the model "By" the different groups- and understand if the differences are true differences or not -similarly to the concept of statistically significant differences in research/ survey data-.

Does this make sense?

Thanks!

Re: Margin of error of predictor screening

Byron_JMP — Wed, 15 Mar 2023 12:46:16 GMT

In JMP Pro I can get the Root Average Square Error (RASE) for In Bag and Out of Bag trees. Using a by group, I can then compare the ratios to see if the model is similarly predictive between groups.

In the case you described, the next step after Predictor Screening is to use the strongest predictors in a regression model (Fit Model) to determine if the predictors' coefficients are significant.

Re: Margin of error of predictor screening

melina_d — Tue, 21 Mar 2023 11:41:17 GMT

Thank you very much Byron, and I'm not sure how I have to run a regression model after the Predictor Screening. Could you please help me out? Thanks!!