A MEASURE OF POST VARIABLE SELECTION ERROR IN MULTIPLE LINEAR REGRESSION, AND ITS ESTIMATION

摘要

Variable selection is frequently one of the first steps in a multiple linear regression analysis. If one follows this with inference using the same data, the effect of the data dependent selection step should be taken into account. Against this background we investigate a measure of post-selection error. Two estimators of this error are proposed, and their behaviour compared in a limited simulation study.