Some Drawbacks of the Simes Test in the Group Sequential Setting

Ajit C. Tamhane; Jiangtao Gou; Alex Dmitrienko

doi:10.1080/19466315.2019.1700156

Some Drawbacks of the Simes Test in the Group Sequential Setting

Ajit C. Tamhane^*, Jiangtao Gou, Alex Dmitrienko

^*Corresponding author for this work

Industrial Engineering and Management Sciences

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

2 Scopus citations

Abstract

The Simes test is designed to test a single null hypothesis formed by taking the intersection of multiple null hypotheses using their p-values. In this short note, we explore a novel application of the Simes test for testing a single null hypothesis in a group sequential setting based on the p-values from sequential looks. We refer to this test as the group sequential Simes test (GSST). It turns out, however, that GSST suffers from some drawbacks. The main drawback is that GSST is uniformly less powerful than the reference group sequential test on which it is based. The reason is that the rejection decision of GSST can be based on a test statistic from an earlier stage, which is not a sufficient statistic. This is also a practical drawback. These drawbacks are discussed in this short note.

Original language	English (US)
Pages (from-to)	390-393
Number of pages	4
Journal	Statistics in Biopharmaceutical Research
Volume	12
Issue number	3
DOIs	https://doi.org/10.1080/19466315.2019.1700156
State	Published - Jul 2 2020

Funding

We gratefully acknowledge the incisive and insightful comments of two referees and an associate editor. One of the referees brought to our attention the articles by Chang (1996) and Jennison and Turnbull (2006).

Keywords

Error spending function
Group sequential procedures
O’Brien and Fleming procedure
Pocock procedure
Simes test

ASJC Scopus subject areas

Statistics and Probability
Pharmaceutical Science

Access to Document

10.1080/19466315.2019.1700156

Cite this

@article{9d7d1401ddaf40aa836c7a0e756515c7,

title = "Some Drawbacks of the Simes Test in the Group Sequential Setting",

abstract = "The Simes test is designed to test a single null hypothesis formed by taking the intersection of multiple null hypotheses using their p-values. In this short note, we explore a novel application of the Simes test for testing a single null hypothesis in a group sequential setting based on the p-values from sequential looks. We refer to this test as the group sequential Simes test (GSST). It turns out, however, that GSST suffers from some drawbacks. The main drawback is that GSST is uniformly less powerful than the reference group sequential test on which it is based. The reason is that the rejection decision of GSST can be based on a test statistic from an earlier stage, which is not a sufficient statistic. This is also a practical drawback. These drawbacks are discussed in this short note.",

keywords = "Error spending function, Group sequential procedures, O{\textquoteright}Brien and Fleming procedure, Pocock procedure, Simes test",

author = "Tamhane, {Ajit C.} and Jiangtao Gou and Alex Dmitrienko",

note = "Publisher Copyright: {\textcopyright} 2020 American Statistical Association.",

year = "2020",

month = jul,

day = "2",

doi = "10.1080/19466315.2019.1700156",

language = "English (US)",

volume = "12",

pages = "390--393",

journal = "Statistics in Biopharmaceutical Research",

issn = "1946-6315",

publisher = "Taylor and Francis Ltd.",

number = "3",

}

TY - JOUR

T1 - Some Drawbacks of the Simes Test in the Group Sequential Setting

AU - Tamhane, Ajit C.

AU - Gou, Jiangtao

AU - Dmitrienko, Alex

PY - 2020/7/2

Y1 - 2020/7/2

N2 - The Simes test is designed to test a single null hypothesis formed by taking the intersection of multiple null hypotheses using their p-values. In this short note, we explore a novel application of the Simes test for testing a single null hypothesis in a group sequential setting based on the p-values from sequential looks. We refer to this test as the group sequential Simes test (GSST). It turns out, however, that GSST suffers from some drawbacks. The main drawback is that GSST is uniformly less powerful than the reference group sequential test on which it is based. The reason is that the rejection decision of GSST can be based on a test statistic from an earlier stage, which is not a sufficient statistic. This is also a practical drawback. These drawbacks are discussed in this short note.

AB - The Simes test is designed to test a single null hypothesis formed by taking the intersection of multiple null hypotheses using their p-values. In this short note, we explore a novel application of the Simes test for testing a single null hypothesis in a group sequential setting based on the p-values from sequential looks. We refer to this test as the group sequential Simes test (GSST). It turns out, however, that GSST suffers from some drawbacks. The main drawback is that GSST is uniformly less powerful than the reference group sequential test on which it is based. The reason is that the rejection decision of GSST can be based on a test statistic from an earlier stage, which is not a sufficient statistic. This is also a practical drawback. These drawbacks are discussed in this short note.

KW - Error spending function

KW - Group sequential procedures

KW - O’Brien and Fleming procedure

KW - Pocock procedure

KW - Simes test

UR - http://www.scopus.com/inward/record.url?scp=85078503634&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85078503634&partnerID=8YFLogxK

U2 - 10.1080/19466315.2019.1700156

DO - 10.1080/19466315.2019.1700156

M3 - Article

AN - SCOPUS:85078503634

SN - 1946-6315

VL - 12

SP - 390

EP - 393

JO - Statistics in Biopharmaceutical Research

JF - Statistics in Biopharmaceutical Research

IS - 3

ER -

Some Drawbacks of the Simes Test in the Group Sequential Setting

Abstract

Funding

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this