A stochastic model for the analysis of bivariate longitudinal AIDS data.

摘要: We present a model for multivariate repeated measures that incorporates random effects, correlated stochastic processes, and measurement errors. The is generalization of the univariate longitudinal data given by Taylor, Cumberland, Sy (1994, Journal American Statistical Association 89, 727-736). process used in this paper integrated Ornstein-Uhlenbeck (OU) process, which includes Brownian motion effects as special limiting cases. This an underlying continuous-time autoregressive order [AR(1)] derivatives observations. allows unequally spaced observations missing values some variables. analyze CD4 T-cell beta-2-microglobulin measurements seroconverters at multiple time points from Los Angeles section Multicenter AIDS Cohort Study. us to investigate relationship between through correlations their serial correlation. suggest follow bivariate process. fit implies increase associated with decrease future but not vice versa, agreeing immunologic postulates about these two