Detecting Serial Correlation in the Error Structure of a Cross-Lagged Panel Model.
Abstract
Cross-lagged panel studies are statistical studies in which two or more variable are measured for a large number of subjects at each of several waves or points in time. The variables divide naturally into two sets and the primary purpose of analysis is to estimate and test the strength of the relationship between the sets. This paper contributes to these studies by developing and applying procedures for detecting the presence of serial correlation in the error structure of the regression models used in such studies. The regression approach was extended by incorporating the cross-effects as parameters in a multivariate regression model and develops procedures to estimate and test these parameters. Both the model with independent errors and the model with serially correlated errors were considered. This paper extends the applicability of this results by considering the problem of serially correlated error structure as opposed to the independent error structure.
Document Details
- Document Type
- Technical Report
- Publication Date
- Feb 02, 1988
- Accession Number
- ADA189070
Entities
People
- Lawrence S. Mayer
- Steven S. Carroll
Organizations
- Stanford University