Abstract
Continuous-variable panel models are widely used in social and business research to assess the relationships between variables over time given measurements at several waves on a single sample of individuals. Emphasis is often placed on estimating and testing the cross-effects which are the regression coefficients indicating the lagged effect of one variable on another. For the simple two-variable model we consider the problem of incorporating the contemporaneous relationship between the variables into the model. Three extensions of the “independent regressions” model are considered. Their similarities and differences are examined. The use of the models is illustrated by examining data on the attitudes toward the criminal justice system and capital punishment for a panel of petit jurors.
Original language | English (US) |
---|---|
Pages (from-to) | 1593-1617 |
Number of pages | 25 |
Journal | Communications in Statistics - Theory and Methods |
Volume | 20 |
Issue number | 5-6 |
DOIs | |
State | Published - Jan 1 1991 |
Keywords
- cross-lagged coefficients
- linear models
- maximum-likelihood estimation
ASJC Scopus subject areas
- Statistics and Probability