Two Guidelines for Bootstrap Hypothesis Testing

Peter Hall, Susan R. Wilson

Biometrics · 1991 · 482 citations · 5 references

Concepts

TL;DR

The paper presents two guidelines for nonparametric bootstrap hypothesis testing. The first guideline advises resampling that reflects the null hypothesis even when the true hypothesis is far from it, while the second recommends using bootstrap methods proven effective in confidence‑interval construction. Violating the first guideline can dramatically reduce test power, especially when the null is far from truth, whereas the second guideline improves level accuracy without affecting power and is useful when test conclusions are equivocal.

Abstract

Two guidelines for nonparametric bootstrap hypothesis testing are highlighted. The first recommends that resampling be done in a way that reflects the null hypothesis, even when the true hypothesis is distant from the null. The second guideline argues that bootstrap hypothesis tests should employ methods that are already recognized as having good features in the closely related problem of confidence interval construction. Violation of the first guideline can seriously reduce the power of a test. Sometimes this reduction is spectacular, since it is most serious when the null hypothesis is grossly in error. The second guideline is of some importance when the conclusion of a test is equivocal. It has no direct bearing on power, but improves the level accuracy of a test.

References

5