Stepwise regression

In <a href="/facts/Statistics/DNPsKGYU">statistics</a>, stepwise regression is a method of fitting <a href="/facts/Regression_model/n6z5Tf7K">regression models</a> 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 <a href="/facts/Explanatory_variable/dINfzIF2">explanatory variables</a> based on some prespecified criterion. Usually, this takes the form of a forward, backward, or combined sequence of <a href="/facts/F-test/Kq8lzNqY">F-tests</a> or <a href="/facts/T-test/dMH8uC28">t-tests</a>.
The frequent practice of fitting the final selected model followed by reporting estimates and confidence intervals without adjusting them to take the model building process into account has led to calls to stop using stepwise model building altogether or to at least make sure model uncertainty is correctly reflected by using prespecified, automatic criteria together with more complex standard error estimates that remain unbiased.

Stepwise regression open-in-new

Stepwise regression