Aarhus University Seal / Aarhus Universitets segl

On the Use of Saddlepoint Approximations in High Dimensional Inference

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaperJournal articleResearchpeer-review

Standard

On the Use of Saddlepoint Approximations in High Dimensional Inference. / Jensen, Jens Ledet.

In: Sankhya A, Vol. 83, No. 1, 02.2021, p. 379-392.

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaperJournal articleResearchpeer-review

Harvard

APA

CBE

MLA

Vancouver

Author

Bibtex

@article{8cb4038f116b4052823e8c868bd5c1ac,
title = "On the Use of Saddlepoint Approximations in High Dimensional Inference",
abstract = "Inference in high dimensional parameter space poses many challenges. One of these is the possible use of saddlepoint approximations. Motivated by a recent use of the saddlepoint approximation to construct a conditional test, we argue that the precision is questionable. We illustrate this by an example giving a 50% relative error in the calculation of the p-value. A power study of the underlying test reveals a low power in many situations. As an alternative it is suggested to use the likelihood ratio test.",
author = "Jensen, {Jens Ledet}",
year = "2021",
month = feb,
doi = "10.1007/s13171-019-00188-x",
language = "English",
volume = "83",
pages = "379--392",
journal = "Sankhya A",
issn = "0976-8378",
publisher = "Springer India",
number = "1",

}

RIS

TY - JOUR

T1 - On the Use of Saddlepoint Approximations in High Dimensional Inference

AU - Jensen, Jens Ledet

PY - 2021/2

Y1 - 2021/2

N2 - Inference in high dimensional parameter space poses many challenges. One of these is the possible use of saddlepoint approximations. Motivated by a recent use of the saddlepoint approximation to construct a conditional test, we argue that the precision is questionable. We illustrate this by an example giving a 50% relative error in the calculation of the p-value. A power study of the underlying test reveals a low power in many situations. As an alternative it is suggested to use the likelihood ratio test.

AB - Inference in high dimensional parameter space poses many challenges. One of these is the possible use of saddlepoint approximations. Motivated by a recent use of the saddlepoint approximation to construct a conditional test, we argue that the precision is questionable. We illustrate this by an example giving a 50% relative error in the calculation of the p-value. A power study of the underlying test reveals a low power in many situations. As an alternative it is suggested to use the likelihood ratio test.

U2 - 10.1007/s13171-019-00188-x

DO - 10.1007/s13171-019-00188-x

M3 - Journal article

VL - 83

SP - 379

EP - 392

JO - Sankhya A

JF - Sankhya A

SN - 0976-8378

IS - 1

ER -