23 January 2021 1 549 Report

When you create a model with RSM, even if the R2 value is large and the result of the analysis of variance is significant, if you look at the individual explanatory variables, if there are many variables that are not significant, is the model unreliable?

I built a model to predict the strength of the alloy from arbitrary fabrication conditions.

The R2 value was 0.78 and the p-value in the analysis of variance for the model was 0.0040, but p > 0.05 for 7 of the 9 explanatory variables.

Similar questions and discussions