Simple estimation of hidden correlation in repeated measures

Thuan Nguyen, Jiming Jiang

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

In medical and social studies, it is often desirable to assess the correlation between characteristics of interest that are not directly observable. In such cases, repeated measures are often available, but the correlation between the repeated measures is not the same as that between the true characteristics that are confounded with the measurement errors. The latter is called the hidden correlation. Previously, the problem has been treated by assuming prior knowledge about the measurement errors or by using relatively complex statistical models, such as the mixed-effects models, with no closed-form expression for the estimated hidden correlation. We propose a simple estimator of the hidden correlation that is very much like the Pearson correlation coefficient, with a closed-form expression, under assumptions much weaker than the mixed-effects model. Simulation results show that the proposed simple estimator performs similarly as the restricted maximum likelihood (REML) estimator in mixed models but is computationally much more efficient than REML. We also made simulation comparison with the Pearson correlation. We considered a real data example. 30 29 20 December 2011 10.1002/sim.4366 Research Article Research Articles

Original languageEnglish (US)
Pages (from-to)3403-3415
Number of pages13
JournalStatistics in Medicine
Volume30
Issue number29
DOIs
StatePublished - Dec 20 2011
Externally publishedYes

Fingerprint

Repeated Measures
Statistical Models
Mixed Effects Model
Pearson Correlation
Research
Measurement Error
Closed-form
Restricted Maximum Likelihood Estimator
Estimator
Restricted Maximum Likelihood
Mixed Model
Prior Knowledge
Correlation coefficient
Statistical Model
Simulation

Keywords

  • Correlation coefficient
  • Hypothesis testing
  • Repeated measures

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability

Cite this

Simple estimation of hidden correlation in repeated measures. / Nguyen, Thuan; Jiang, Jiming.

In: Statistics in Medicine, Vol. 30, No. 29, 20.12.2011, p. 3403-3415.

Research output: Contribution to journalArticle

Nguyen, Thuan ; Jiang, Jiming. / Simple estimation of hidden correlation in repeated measures. In: Statistics in Medicine. 2011 ; Vol. 30, No. 29. pp. 3403-3415.
@article{e8f845cccc6f4f848a341e99e5c9f895,
title = "Simple estimation of hidden correlation in repeated measures",
abstract = "In medical and social studies, it is often desirable to assess the correlation between characteristics of interest that are not directly observable. In such cases, repeated measures are often available, but the correlation between the repeated measures is not the same as that between the true characteristics that are confounded with the measurement errors. The latter is called the hidden correlation. Previously, the problem has been treated by assuming prior knowledge about the measurement errors or by using relatively complex statistical models, such as the mixed-effects models, with no closed-form expression for the estimated hidden correlation. We propose a simple estimator of the hidden correlation that is very much like the Pearson correlation coefficient, with a closed-form expression, under assumptions much weaker than the mixed-effects model. Simulation results show that the proposed simple estimator performs similarly as the restricted maximum likelihood (REML) estimator in mixed models but is computationally much more efficient than REML. We also made simulation comparison with the Pearson correlation. We considered a real data example. 30 29 20 December 2011 10.1002/sim.4366 Research Article Research Articles",
keywords = "Correlation coefficient, Hypothesis testing, Repeated measures",
author = "Thuan Nguyen and Jiming Jiang",
year = "2011",
month = "12",
day = "20",
doi = "10.1002/sim.4366",
language = "English (US)",
volume = "30",
pages = "3403--3415",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "29",

}

TY - JOUR

T1 - Simple estimation of hidden correlation in repeated measures

AU - Nguyen, Thuan

AU - Jiang, Jiming

PY - 2011/12/20

Y1 - 2011/12/20

N2 - In medical and social studies, it is often desirable to assess the correlation between characteristics of interest that are not directly observable. In such cases, repeated measures are often available, but the correlation between the repeated measures is not the same as that between the true characteristics that are confounded with the measurement errors. The latter is called the hidden correlation. Previously, the problem has been treated by assuming prior knowledge about the measurement errors or by using relatively complex statistical models, such as the mixed-effects models, with no closed-form expression for the estimated hidden correlation. We propose a simple estimator of the hidden correlation that is very much like the Pearson correlation coefficient, with a closed-form expression, under assumptions much weaker than the mixed-effects model. Simulation results show that the proposed simple estimator performs similarly as the restricted maximum likelihood (REML) estimator in mixed models but is computationally much more efficient than REML. We also made simulation comparison with the Pearson correlation. We considered a real data example. 30 29 20 December 2011 10.1002/sim.4366 Research Article Research Articles

AB - In medical and social studies, it is often desirable to assess the correlation between characteristics of interest that are not directly observable. In such cases, repeated measures are often available, but the correlation between the repeated measures is not the same as that between the true characteristics that are confounded with the measurement errors. The latter is called the hidden correlation. Previously, the problem has been treated by assuming prior knowledge about the measurement errors or by using relatively complex statistical models, such as the mixed-effects models, with no closed-form expression for the estimated hidden correlation. We propose a simple estimator of the hidden correlation that is very much like the Pearson correlation coefficient, with a closed-form expression, under assumptions much weaker than the mixed-effects model. Simulation results show that the proposed simple estimator performs similarly as the restricted maximum likelihood (REML) estimator in mixed models but is computationally much more efficient than REML. We also made simulation comparison with the Pearson correlation. We considered a real data example. 30 29 20 December 2011 10.1002/sim.4366 Research Article Research Articles

KW - Correlation coefficient

KW - Hypothesis testing

KW - Repeated measures

UR - http://www.scopus.com/inward/record.url?scp=82155166336&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=82155166336&partnerID=8YFLogxK

U2 - 10.1002/sim.4366

DO - 10.1002/sim.4366

M3 - Article

C2 - 21997471

AN - SCOPUS:82155166336

VL - 30

SP - 3403

EP - 3415

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 29

ER -