Second-order correctness of the Poisson bootstrap

G. Jogesh Babu, P. K. Pathak, C. R. Rao

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

Rao, Pathak and Koltchinskii have recently studied a sequential approach to resampling in which resampling is carried out sequentially one-by-one (with replacement each time) until the bootstrap sample contains m ≈ (1 - e -1)n ≈ 0.632n distinct observations from the original sample. In our previous work, we have established that the main empirical characteristics of the sequential bootstrap go through, in the sense of being within a distance O(n-3/4) from those of the usual bootstrap. However, the theoretical justification of the second-order correctness of the sequential bootstrap is somewhat difficult. It is the main topic of this investigation. Among other things, we accomplish it by approximating our sequential scheme by a resampling scheme based on the Poisson distribution with mean μ = 1 and censored at X = 0.

Original languageEnglish (US)
Pages (from-to)1666-1683
Number of pages18
JournalAnnals of Statistics
Volume27
Issue number5
StatePublished - Oct 1999

Fingerprint

Bootstrap
Correctness
Siméon Denis Poisson
Resampling
Poisson distribution
Justification
Thing
Replacement
Distinct

All Science Journal Classification (ASJC) codes

  • Mathematics(all)
  • Statistics and Probability

Cite this

Babu, G. J., Pathak, P. K., & Rao, C. R. (1999). Second-order correctness of the Poisson bootstrap. Annals of Statistics, 27(5), 1666-1683.
Babu, G. Jogesh ; Pathak, P. K. ; Rao, C. R. / Second-order correctness of the Poisson bootstrap. In: Annals of Statistics. 1999 ; Vol. 27, No. 5. pp. 1666-1683.
@article{7f0324cb19bd4f2280e3bd6e9f0a0845,
title = "Second-order correctness of the Poisson bootstrap",
abstract = "Rao, Pathak and Koltchinskii have recently studied a sequential approach to resampling in which resampling is carried out sequentially one-by-one (with replacement each time) until the bootstrap sample contains m ≈ (1 - e -1)n ≈ 0.632n distinct observations from the original sample. In our previous work, we have established that the main empirical characteristics of the sequential bootstrap go through, in the sense of being within a distance O(n-3/4) from those of the usual bootstrap. However, the theoretical justification of the second-order correctness of the sequential bootstrap is somewhat difficult. It is the main topic of this investigation. Among other things, we accomplish it by approximating our sequential scheme by a resampling scheme based on the Poisson distribution with mean μ = 1 and censored at X = 0.",
author = "Babu, {G. Jogesh} and Pathak, {P. K.} and Rao, {C. R.}",
year = "1999",
month = "10",
language = "English (US)",
volume = "27",
pages = "1666--1683",
journal = "Annals of Statistics",
issn = "0090-5364",
publisher = "Institute of Mathematical Statistics",
number = "5",

}

Babu, GJ, Pathak, PK & Rao, CR 1999, 'Second-order correctness of the Poisson bootstrap', Annals of Statistics, vol. 27, no. 5, pp. 1666-1683.

Second-order correctness of the Poisson bootstrap. / Babu, G. Jogesh; Pathak, P. K.; Rao, C. R.

In: Annals of Statistics, Vol. 27, No. 5, 10.1999, p. 1666-1683.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Second-order correctness of the Poisson bootstrap

AU - Babu, G. Jogesh

AU - Pathak, P. K.

AU - Rao, C. R.

PY - 1999/10

Y1 - 1999/10

N2 - Rao, Pathak and Koltchinskii have recently studied a sequential approach to resampling in which resampling is carried out sequentially one-by-one (with replacement each time) until the bootstrap sample contains m ≈ (1 - e -1)n ≈ 0.632n distinct observations from the original sample. In our previous work, we have established that the main empirical characteristics of the sequential bootstrap go through, in the sense of being within a distance O(n-3/4) from those of the usual bootstrap. However, the theoretical justification of the second-order correctness of the sequential bootstrap is somewhat difficult. It is the main topic of this investigation. Among other things, we accomplish it by approximating our sequential scheme by a resampling scheme based on the Poisson distribution with mean μ = 1 and censored at X = 0.

AB - Rao, Pathak and Koltchinskii have recently studied a sequential approach to resampling in which resampling is carried out sequentially one-by-one (with replacement each time) until the bootstrap sample contains m ≈ (1 - e -1)n ≈ 0.632n distinct observations from the original sample. In our previous work, we have established that the main empirical characteristics of the sequential bootstrap go through, in the sense of being within a distance O(n-3/4) from those of the usual bootstrap. However, the theoretical justification of the second-order correctness of the sequential bootstrap is somewhat difficult. It is the main topic of this investigation. Among other things, we accomplish it by approximating our sequential scheme by a resampling scheme based on the Poisson distribution with mean μ = 1 and censored at X = 0.

UR - http://www.scopus.com/inward/record.url?scp=0033212082&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033212082&partnerID=8YFLogxK

M3 - Article

VL - 27

SP - 1666

EP - 1683

JO - Annals of Statistics

JF - Annals of Statistics

SN - 0090-5364

IS - 5

ER -