A method for comparing two normal means using combined samples of correlated and uncorrelated data

Stephen W. Looney, Peter W. Jones

Research output: Contribution to journalArticle

20 Citations (Scopus)

Abstract

A method is proposed for comparing two normal means when the data consist of some observations that are correlated and others that are uncorrelated. Typically, such data will consist of one subsample in which the observations for treatment 1 and treatment 2 are independent of each other, and another subsample which consists of paired observations taken under both treatments. The proposed method is developed using asymptotic results and is evaluated using simulation. The simulation results indicate that the proposed method can provide substantial improvement in type I error rate and power when compared with standard methods of analysis.

Original languageEnglish (US)
Pages (from-to)1601-1610
Number of pages10
JournalStatistics in Medicine
Volume22
Issue number9
DOIs
StatePublished - May 15 2003

Fingerprint

Type I Error Rate
Simulation
Observation
Standards

Keywords

  • Independence
  • Paired t-test
  • Power
  • Simulation
  • Two-sample problem
  • t-test

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability

Cite this

A method for comparing two normal means using combined samples of correlated and uncorrelated data. / Looney, Stephen W.; Jones, Peter W.

In: Statistics in Medicine, Vol. 22, No. 9, 15.05.2003, p. 1601-1610.

Research output: Contribution to journalArticle

@article{5f62470ae0834ce3b17c42090030747e,
title = "A method for comparing two normal means using combined samples of correlated and uncorrelated data",
abstract = "A method is proposed for comparing two normal means when the data consist of some observations that are correlated and others that are uncorrelated. Typically, such data will consist of one subsample in which the observations for treatment 1 and treatment 2 are independent of each other, and another subsample which consists of paired observations taken under both treatments. The proposed method is developed using asymptotic results and is evaluated using simulation. The simulation results indicate that the proposed method can provide substantial improvement in type I error rate and power when compared with standard methods of analysis.",
keywords = "Independence, Paired t-test, Power, Simulation, Two-sample problem, t-test",
author = "Looney, {Stephen W.} and Jones, {Peter W.}",
year = "2003",
month = "5",
day = "15",
doi = "10.1002/sim.1514",
language = "English (US)",
volume = "22",
pages = "1601--1610",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "9",

}

TY - JOUR

T1 - A method for comparing two normal means using combined samples of correlated and uncorrelated data

AU - Looney, Stephen W.

AU - Jones, Peter W.

PY - 2003/5/15

Y1 - 2003/5/15

N2 - A method is proposed for comparing two normal means when the data consist of some observations that are correlated and others that are uncorrelated. Typically, such data will consist of one subsample in which the observations for treatment 1 and treatment 2 are independent of each other, and another subsample which consists of paired observations taken under both treatments. The proposed method is developed using asymptotic results and is evaluated using simulation. The simulation results indicate that the proposed method can provide substantial improvement in type I error rate and power when compared with standard methods of analysis.

AB - A method is proposed for comparing two normal means when the data consist of some observations that are correlated and others that are uncorrelated. Typically, such data will consist of one subsample in which the observations for treatment 1 and treatment 2 are independent of each other, and another subsample which consists of paired observations taken under both treatments. The proposed method is developed using asymptotic results and is evaluated using simulation. The simulation results indicate that the proposed method can provide substantial improvement in type I error rate and power when compared with standard methods of analysis.

KW - Independence

KW - Paired t-test

KW - Power

KW - Simulation

KW - Two-sample problem

KW - t-test

UR - http://www.scopus.com/inward/record.url?scp=0037687727&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0037687727&partnerID=8YFLogxK

U2 - 10.1002/sim.1514

DO - 10.1002/sim.1514

M3 - Article

C2 - 12704618

AN - SCOPUS:0037687727

VL - 22

SP - 1601

EP - 1610

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 9

ER -