Abstract:
|
A principal aim of current psychiatric research is to chart the trajectories of mental illness, in part by studying patterns of brain maturation. This talk will examine two statistical issues in this domain, each playing a role that is potentially crucial but poorly understood. First, we discuss the relative efficiency of cross-sectional and longitudinal designs when the "trajectory" of interest is the mean of some quantity, such as cortical thickness, as a function of age. A classical variance inflation factor is generalized from the estimation of a scalar to the present setting of function estimation, and is further extended to penalized smoothing. Second, we consider the use of functional principal component analysis (FPCA) for estimation of individual trajectories; specifically we show how penalized smoothing of the covariance surface can have surprising effects on the results. The ideas are illustrated with data from a large longitudinal study of cortical development.
|