Notwithstanding the issues with certain variable selection procedures like stepwise regression, suppose you run some sort of variable selection to find significant categorical covariates in a regression. Then, you merge categories in order to increase the precision of your estimates.

Could this possibly be problematic, since the new coding may make the significant variable from the variable selection procedure insignificant?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.