3
$\begingroup$

In a classic paper on conducting bootstrapped hypothesis tests, Hall & Wilson (1991) present the following guideline, which I am trying to understand:

enter image description here

On first read, this seemed obvious: $\widehat{\theta}^{*}$ are being resampled under $H_0$, so are consistent for $\theta_0$, whereas $\widehat{\theta}$ is consistent for $\theta$. Thus, $\widehat{\theta}^{*} - \widehat{\theta}$ is consistent for $\theta_0 - \theta$, which makes sense. In contrast, $\widehat{\theta}^{*} - \theta_0$ is consistent for $\theta_0 - \theta_0 = 0$, which is not useful.

But then I realized, on rereading the paragraph above the First Guideline and a subsequent example, that Hall & Wilson are computing $\widehat{\theta}^{*}$ by resampling directly from the sample, rather than under the null. In that case, I have no idea why they recommend resampling $\widehat{\theta}^{*} - \widehat{\theta}$, which is consistent for 0!

References

Hall, P., & Wilson, S. R. (1991). Two guidelines for bootstrap hypothesis testing. Biometrics, 757-762.

|cite|improve this question
$\endgroup$
3
$\begingroup$

The idea is that irrespective of the choice of $\theta_0$ the bootstrap distribution of $\widehat\theta^*$ - $\widehat\theta$ resembles the sampling distribution of $\widehat\theta$ - $\theta$. If you instead use $\theta_0$ you will be pulling the estimate closer to $\theta_0$ and thus lose power under the alternative. I think this is essentially what Hall and Wilson are saying but possibly a little simpler and hopefully clearer for you.

|cite|improve this answer
$\endgroup$
  • $\begingroup$ Thanks -- I think I understand. Is it correct to say that this algorithm allows us to conduct a hypothesis test under $H_0$ while avoiding the difficulties of resampling under $H_0$ (because we're just resampling from the original)? $\endgroup$ – half-pass Jul 7 '17 at 2:42
  • 1
    $\begingroup$ That sounds like a reasonable statement to me. $\endgroup$ – Michael R. Chernick Jul 7 '17 at 2:44
  • $\begingroup$ Interesting. (As an aside, your textbook on bootstrapping is wonderful -- thank you!) $\endgroup$ – half-pass Jul 7 '17 at 2:56
  • $\begingroup$ Thanks for the compliment. I briefly discuss the Hall-Wilson paper in my book. You may also like to look at Peter Hall's book which also covers this hypothesis testing issue. $\endgroup$ – Michael R. Chernick Jul 7 '17 at 3:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.