a method of paring down the a set of candidate predictor variables in a linear model one at a time until all variables left in the model are useful in explaining variability in the response, y