Using a serial marker to predict a repeated measures outcome in a cohort design--results of a simulation study.

摘要: Abstract Consider the cohort design and suppose that outcome of primary interest is a continuous random variable observed repeatedly over time. Suppose there second clinical relevance which also repeatedly. We are interested in assessing whether “serial marker” some sense predictive outcome. would like to predict trend for assuming marker follows profile specific interest. In series earlier papers, we have addressed these issues by applying bivariate repeated measures model. One regression model was prescribed relate important explanatory variables, while serial marker. this paper, perform simulation studies investigate empirical properties approach. Bivariate data were generated at random, basic study parameters including sample size, numb...