Search Results

Title: Best of Both Worlds: Combining Autoregressive and Latent Curve Models
Resulting in 1 citation.
1. Curran, Patrick J.
Bollen, Kenneth A.
Best of Both Worlds: Combining Autoregressive and Latent Curve Models
In: New Methods of the Analysis of Change. A. G. Sayer, ed. Washington, DC: American Psychological Association, 2001: pp. 107-135
Cohort(s): Children of the NLSY79
Publisher: American Psychological Association (APA)
Keyword(s): Behavior Problems Index (BPI); Depression (see also CESD); Markov chain / Markov model; Modeling, Fixed Effects; Modeling, Growth Curve/Latent Trajectory Analysis; Statistical Analysis

Discusses the autoregressive model (or "fixed effects Markov simplex model") and random coefficient growth curve models as being two analytic approaches to the theoretical conceptualization and statistical analysis of panel data. An extended empirical example is presented in order to illustrate the authors' ongoing efforts to synthesize these two models. They begin with a description of a theoretical substantive question that motivates the development of the synthesized model, they then present a review of the univariate and bivariate autoregressive simplex models followed by a general description of the univariate and bivariate latent curve models. The synthesis of the simplex and latent curve models is proposed for both the univariate and bivariate cases, and these are applied to the empirical data set to evaluate a series of questions relating to the developmental relation between antisocial behavior and depressive symptomatology.
Bibliography Citation
Curran, Patrick J. and Kenneth A. Bollen. "Best of Both Worlds: Combining Autoregressive and Latent Curve Models" In: New Methods of the Analysis of Change. A. G. Sayer, ed. Washington, DC: American Psychological Association, 2001: pp. 107-135