Conditional saddlepoint approximations for non-continuous and non-lattice distributions

John E. Kolassa; John Robinson

doi:10.1016/j.jspi.2005.11.003

Conditional saddlepoint approximations for non-continuous and non-lattice distributions

John E. Kolassa, John Robinson

Research output: Contribution to journal › Article › peer-review

1 Scopus citations

Abstract

This manuscript presents an approximation to the distribution function of a smooth transformation of a random vector, conditional on the event that values of other smooth transformations of the same random vector lie in a small rectangle. This approximation is used to extend the application of standard saddlepoint conditional tail area approximations in circumstances beyond continuous and lattice cases currently justified in the literature. We consider application to two examples, finite sampling and score testing in logistic regression, where conditioning on a rectangle is essential.

Original language	English (US)
Pages (from-to)	133-147
Number of pages	15
Journal	Journal of Statistical Planning and Inference
Volume	137
Issue number	1
DOIs	https://doi.org/10.1016/j.jspi.2005.11.003
State	Published - Jan 1 2007

All Science Journal Classification (ASJC) codes

Statistics and Probability
Statistics, Probability and Uncertainty
Applied Mathematics

Keywords

Conditional inference
Saddlepoint approximation

Access to Document

10.1016/j.jspi.2005.11.003

Cite this

@article{d9cbf9ca51f747cca37506fc550fee58,

title = "Conditional saddlepoint approximations for non-continuous and non-lattice distributions",

abstract = "This manuscript presents an approximation to the distribution function of a smooth transformation of a random vector, conditional on the event that values of other smooth transformations of the same random vector lie in a small rectangle. This approximation is used to extend the application of standard saddlepoint conditional tail area approximations in circumstances beyond continuous and lattice cases currently justified in the literature. We consider application to two examples, finite sampling and score testing in logistic regression, where conditioning on a rectangle is essential.",

keywords = "Conditional inference, Saddlepoint approximation",

author = "Kolassa, {John E.} and John Robinson",

note = "Funding Information: This research was supported by ARC Discovery Project DP0451722 and by US NSF DMS 0505499. The authors gratefully acknowledge assistance from an anonymous referee and Lars Nordmann.",

year = "2007",

month = jan,

day = "1",

doi = "10.1016/j.jspi.2005.11.003",

language = "English (US)",

volume = "137",

pages = "133--147",

journal = "Journal of Statistical Planning and Inference",

issn = "0378-3758",

publisher = "Elsevier B.V.",

number = "1",

}

TY - JOUR

T1 - Conditional saddlepoint approximations for non-continuous and non-lattice distributions

AU - Kolassa, John E.

AU - Robinson, John

N1 - Funding Information: This research was supported by ARC Discovery Project DP0451722 and by US NSF DMS 0505499. The authors gratefully acknowledge assistance from an anonymous referee and Lars Nordmann.

PY - 2007/1/1

Y1 - 2007/1/1

N2 - This manuscript presents an approximation to the distribution function of a smooth transformation of a random vector, conditional on the event that values of other smooth transformations of the same random vector lie in a small rectangle. This approximation is used to extend the application of standard saddlepoint conditional tail area approximations in circumstances beyond continuous and lattice cases currently justified in the literature. We consider application to two examples, finite sampling and score testing in logistic regression, where conditioning on a rectangle is essential.

AB - This manuscript presents an approximation to the distribution function of a smooth transformation of a random vector, conditional on the event that values of other smooth transformations of the same random vector lie in a small rectangle. This approximation is used to extend the application of standard saddlepoint conditional tail area approximations in circumstances beyond continuous and lattice cases currently justified in the literature. We consider application to two examples, finite sampling and score testing in logistic regression, where conditioning on a rectangle is essential.

KW - Conditional inference

KW - Saddlepoint approximation

UR - http://www.scopus.com/inward/record.url?scp=33748414463&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33748414463&partnerID=8YFLogxK

U2 - 10.1016/j.jspi.2005.11.003

DO - 10.1016/j.jspi.2005.11.003

M3 - Article

AN - SCOPUS:33748414463

SN - 0378-3758

VL - 137

SP - 133

EP - 147

JO - Journal of Statistical Planning and Inference

JF - Journal of Statistical Planning and Inference

IS - 1

ER -

Conditional saddlepoint approximations for non-continuous and non-lattice distributions

Abstract

All Science Journal Classification (ASJC) codes

Keywords

Access to Document

Other files and links

Fingerprint

Cite this