The Size Distortion of Bootstrap Tests

Preprint
Publisher Website
Google Scholar
    • preprint
      • Published in RePEc
Abstract
Bootstrap tests are tests for which the signicance level is calculated by some sort of bootstrap procedure which may be parametric or nonparametric We provide a theoretical framework in which to study the size distortions of bootstrap P values We show that in many circumstances the size distortion of a bootstrap test will be one whole order of magnitude smaller than that of the corresponding asymptotic test We also show that at least in the parametric case the magnitude of the distortion will depend on the shape of what we call the P value function Monte Carlo results are presented for the case of nonnested hypothesis tests These results conrm and illustrate the utility of our theoretical results and they also suggest that bootstrap tests may often work extremely well in practice This research was supported in part by grants from the Social Sciences and Humanities Research Council of Canada An earlier version was presented at Universidad Carlos III de Madrid Universidad Complutense de Madrid Cambridge University INSEECREST Paris CORE LouvainlaNeuve the Tinbergen Institute Amsterdam the University of Geneva the European University Institute Florence the ESRC Econometrics Conference Bristol and the Berkeley Symposium on the Bootstrap We are grateful to many seminar participants and to two anonymous referees for comments on the earlier version We are especially grateful to Joel Horowitz not only for comments but also for his probing questions that led us to clarify the paper The paper was written while the second author was visiting GREQAM
All Related Versions

This publication has 0 references indexed in Scilit: