×

On the relative accuracy of certain bootstrap procedures. (English) Zbl 0941.62053

Summary: We show the second-order relative accuracy, on bounded sets, of the Studentized bootstrap, exponentially tilted bootstrap and nonparametric likelihood tilted bootstrap, for means and smooth functions of means. We also consider the relative errors for larger deviations. Our method exploits certain connections between Edgeworth and saddlepoint approximations to simplify the computations.

MSC:

62G09 Nonparametric statistical resampling methods
62G20 Asymptotic properties of nonparametric inference
62G10 Nonparametric hypothesis testing

References:

[1] Barndorff-Nielsen, Inference based on full or partial parameters based on the standardized signed log likelihood ratio, Biometrika 73 pp 307– (1986) · Zbl 0605.62020
[2] Becker, The New S Language (1988)
[3] Daniels, Saddlepoint approximation for the studentized mean, with an application to the bootstrap, Biometrika 78 pp 169– (1991)
[4] Davison, Saddlepoint approximations in resampling methods, Biometrika 75 pp 417– (1988) · Zbl 0651.62018
[5] DiCiccio, Nonparametric confidence limits by resampling methods and least favourable families, Internal. Statist. Rev. 58 pp 59– (1990) · Zbl 0715.62090
[6] Feuerverger, On the second order relative accuracy of certain bootstrap and saddlepoint approximation procedures (1998)
[7] Hall, Theoretical comparison of bootstrap confidence intervals (with discussion), Ann. Statist. 16 pp 927– (1988)
[8] Hall, The Bootstrap and Edgeworth Expansion (1992) · Zbl 0744.62026 · doi:10.1007/978-1-4612-4384-7
[9] Jing, On the bootstrap saddlepoint approximations, Biometrika 81 pp 211– (1994) · Zbl 0796.62015
[10] Jing, Saddlepoint approximations for marginal and conditional probabilities of transformed variables, Ann. Statist. 22 pp 1115– (1994) · Zbl 0817.62007
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.