Abstract:
|
In studying the pathway to obesity, a variation of the multiple indicators, multiple cause measurement error (MIMIC ME) models can be applied. In this paper, we define the functional MIMIC ME model by extending the linear MIMIC ME model to allow longitudinal responses that appear as curves over a given period of time. The mean curves are modeled using basis splines and functional principal components. We also propose a novel approach to identifying the classical measurement error associated with approximating true metabolic rate by heat production based on functional principal components. In addition to defining the FMIMIC ME model, we estimate the model parameters using the EM algorithm. We also provide a discussion of the model's identifiability. The model is applied to study the relationship between metabolic rate and the multiple indicators of energy expenditure among Zucker diabetic fatty rats. Results from a brief simulation study are also provided.
|