| Abstract: | Although the concept of the sampling distribution is at the core of much of what we do in econometrics, it is a concept that is often difficult for students to grasp. The thought process behind bootstrapping provides a way for students to conceptualize the sampling distribution in a way that is intuitive and visual. However, teaching students to write bootstrapping code on their own would come at great expense in terms of instruction time. The author proposes the use of instructor-written macros as a balance between these opposing interests, and he focuses on the use of the nonparametric “pairwise” bootstrap procedure in order to make an intuitive link between bootstrapping and sampling from a population (one of the first ideas students meet in statistics). |
