Better saddlepoint confidence intervals via bootstrap calibration

Zheng, Xiaodong

doi:10.1090/S0002-9939-98-04417-7

Better saddlepoint confidence intervals via bootstrap calibration
HTML articles powered by AMS MathViewer

by Xiaodong Zheng PDF

Proc. Amer. Math. Soc. 126 (1998), 3669-3679 Request permission

Abstract:

Confidence interval construction for parameters of lattice distributions is considered. By using saddlepoint formulas and bootstrap calibration, we obtain relatively short intervals and bounds with $O(n^{-3/2})$ coverage errors, in contrast with $O(n^{-1})$ and $O(n^{-1/2})$ coverage errors for normal theory intervals and bounds when the population distribution is absolutely continuous. Closed form solutions are also provided for the cases of binomial and Poisson distributions. The method is illustrated by some simulation results.

References

Blyth, C.R. (1986). Approximate binomial confidence limits. J. Amer. Statist. Assoc. 81 843–855; Corridenda (1989), 84, 636.
H. E. Daniels, Tail probability approximations, Internat. Statist. Rev. 55 (1987), no. 1, 37–48 (English, with French summary). MR 962940, DOI 10.2307/1403269
Peter Hall, Improving the normal approximation when constructing one-sided confidence intervals for binomial or Poisson parameters, Biometrika 69 (1982), no. 3, 647–652. MR 695211, DOI 10.1093/biomet/69.3.647
Hall, P. (1987). On the bootstrap and continuity correction. J. R. Statist. Soc. B 49 82–89.
Peter Hall, Theoretical comparison of bootstrap confidence intervals, Ann. Statist. 16 (1988), no. 3, 927–985. With a discussion and a reply by the author. MR 959185, DOI 10.1214/aos/1176350933
Erich Leo Lehmann, Theory of point estimation, Wiley Series in Probability and Mathematical Statistics: Probability and Mathematical Statistics, John Wiley & Sons, Inc., New York, 1983. A Wiley Publication in Mathematical Statistics. MR 702834, DOI 10.1007/978-1-4757-2769-2
E. L. Lehmann, Testing statistical hypotheses, 2nd ed., Wiley Series in Probability and Mathematical Statistics: Probability and Mathematical Statistics, John Wiley & Sons, Inc., New York, 1986. MR 852406, DOI 10.1007/978-1-4757-1923-9
Wei-Yin Loh, Calibrating confidence coefficients, J. Amer. Statist. Assoc. 82 (1987), no. 397, 155–162. MR 883343, DOI 10.1080/01621459.1987.10478408
Loh, W.-Y. (1988). Discussion of “Theoretical comparison of bootstrap confidence intervals” by P. Hall. Ann. Statist. 16 972–976.
Loh, W.-Y. (1991). Bootstrap calibration for confidence interval construction and selection. Statistica Sinica 1 477–491.
Robert Lugannani and Stephen Rice, Saddle point approximation for the distribution of the sum of independent random variables, Adv. in Appl. Probab. 12 (1980), no. 2, 475–490. MR 569438, DOI 10.2307/1426607
Kesar Singh, On the asymptotic accuracy of Efron’s bootstrap, Ann. Statist. 9 (1981), no. 6, 1187–1195. MR 630102
Maureen Tingley and Christopher Field, Small-sample confidence intervals, J. Amer. Statist. Assoc. 85 (1990), no. 410, 427–434. MR 1141743, DOI 10.1080/01621459.1990.10476216
Xiaodong Zheng and Wei-Yin Loh, Bootstrapping binomial confidence intervals, J. Statist. Plann. Inference 43 (1995), no. 3, 355–380. MR 1323066, DOI 10.1016/0378-3758(94)00031-P