Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I am currently building my model and I have chosen to use a step-wise backwards elimination method. I have many categorical variables and am using the "CLASS" statement in SAS within my regression model. How do I determine if a categorical variable should stay in? Is it based on the TYPE III Analysis of Effects p-value? or is it based on the P for trend value?

share|improve this question
Step-wise model selection methods are not recommended for either continuous or categorical variables. To understand why, you may find it helpful to read my answer here: algorithms-for-automatic-model-selection. – gung Dec 14 '12 at 21:46
You can use penalized methods such as LASSO or LAR by using PROC GLMSELECT – Peter Flom Dec 14 '12 at 22:18
See also my answer as to why to abandon stepwise approaches to model building here. – Alexis Feb 13 '15 at 16:11

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.