Stepwise Regression

Stepwise regression is a method of fitting regression models in which the choice of predictive variables is carried out by an automatic procedure. In each step, a variable is considered for addition to or subtraction from the set of explanatory variables based on some prespecified criterion. The procedure is carried out until no further improvement is possible.

This procedure is often called hierarchical regression when the order of entry of the variables is predetermined. The stepwise regression procedure is a generalization of the hierarchical regression procedure. Unfortunately, hierarchical regression is sometimes also used to refer to multilevel modeling or mixed-effects modeling, which is a different concept.