Analysis of the cross-effects in a cross-lagged panel study

Steven S. Carroll, Lawrence S. Mayer

Research output: Contribution to journalArticle

Abstract

Panel studies are statistical studies in which two or more variables are observed for two or more subjects at two or more points in time, Cross-lagged panel studies are comprised of continuous variables which divide naturally into two sets, and often the primary statistical issue is to estimate and test the cross-effects which indicate the degree to which each set is related to the other over time. By taking a regression approach to modeling the relationships, we apply multivariate regression methodology to make inferences about the regression coefficients in a cross-lagged panel model. In particular we develop a test of the hypothesis that the regression coefficients indicating the cross-effects are equal and develop simultaneous confidence bounds for various linear combinations of these regression coefficients.

Original languageEnglish (US)
Pages (from-to)3379-3399
Number of pages21
JournalCommunications in Statistics - Theory and Methods
Volume15
Issue number11
DOIs
StatePublished - Jan 1 1986

Fingerprint

Regression Coefficient
Confidence Bounds
Multivariate Regression
Continuous Variables
Divides
Linear Combination
Regression
Methodology
Modeling
Estimate
Model

Keywords

  • Cross-lagged
  • Multivariate regression
  • Panel analysis

ASJC Scopus subject areas

  • Statistics and Probability

Cite this

Analysis of the cross-effects in a cross-lagged panel study. / Carroll, Steven S.; Mayer, Lawrence S.

In: Communications in Statistics - Theory and Methods, Vol. 15, No. 11, 01.01.1986, p. 3379-3399.

Research output: Contribution to journalArticle

Carroll, Steven S. ; Mayer, Lawrence S. / Analysis of the cross-effects in a cross-lagged panel study. In: Communications in Statistics - Theory and Methods. 1986 ; Vol. 15, No. 11. pp. 3379-3399.
@article{a6eefad6ab754c2db271b99877101a22,
title = "Analysis of the cross-effects in a cross-lagged panel study",
abstract = "Panel studies are statistical studies in which two or more variables are observed for two or more subjects at two or more points in time, Cross-lagged panel studies are comprised of continuous variables which divide naturally into two sets, and often the primary statistical issue is to estimate and test the cross-effects which indicate the degree to which each set is related to the other over time. By taking a regression approach to modeling the relationships, we apply multivariate regression methodology to make inferences about the regression coefficients in a cross-lagged panel model. In particular we develop a test of the hypothesis that the regression coefficients indicating the cross-effects are equal and develop simultaneous confidence bounds for various linear combinations of these regression coefficients.",
keywords = "Cross-lagged, Multivariate regression, Panel analysis",
author = "Carroll, {Steven S.} and Mayer, {Lawrence S.}",
year = "1986",
month = "1",
day = "1",
doi = "10.1080/03610928608829316",
language = "English (US)",
volume = "15",
pages = "3379--3399",
journal = "Communications in Statistics - Theory and Methods",
issn = "0361-0926",
publisher = "Taylor and Francis Ltd.",
number = "11",

}

TY - JOUR

T1 - Analysis of the cross-effects in a cross-lagged panel study

AU - Carroll, Steven S.

AU - Mayer, Lawrence S.

PY - 1986/1/1

Y1 - 1986/1/1

N2 - Panel studies are statistical studies in which two or more variables are observed for two or more subjects at two or more points in time, Cross-lagged panel studies are comprised of continuous variables which divide naturally into two sets, and often the primary statistical issue is to estimate and test the cross-effects which indicate the degree to which each set is related to the other over time. By taking a regression approach to modeling the relationships, we apply multivariate regression methodology to make inferences about the regression coefficients in a cross-lagged panel model. In particular we develop a test of the hypothesis that the regression coefficients indicating the cross-effects are equal and develop simultaneous confidence bounds for various linear combinations of these regression coefficients.

AB - Panel studies are statistical studies in which two or more variables are observed for two or more subjects at two or more points in time, Cross-lagged panel studies are comprised of continuous variables which divide naturally into two sets, and often the primary statistical issue is to estimate and test the cross-effects which indicate the degree to which each set is related to the other over time. By taking a regression approach to modeling the relationships, we apply multivariate regression methodology to make inferences about the regression coefficients in a cross-lagged panel model. In particular we develop a test of the hypothesis that the regression coefficients indicating the cross-effects are equal and develop simultaneous confidence bounds for various linear combinations of these regression coefficients.

KW - Cross-lagged

KW - Multivariate regression

KW - Panel analysis

UR - http://www.scopus.com/inward/record.url?scp=84953574707&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84953574707&partnerID=8YFLogxK

U2 - 10.1080/03610928608829316

DO - 10.1080/03610928608829316

M3 - Article

AN - SCOPUS:84953574707

VL - 15

SP - 3379

EP - 3399

JO - Communications in Statistics - Theory and Methods

JF - Communications in Statistics - Theory and Methods

SN - 0361-0926

IS - 11

ER -