We have diverse interest in solving methodological issues in statistics. Our past and present statistical projects include diagnostic measures, stochastic approximation algorithm, structural equation models, mixed effect models, spline regression, missing data problems, variable selections, empirical likelihood, mixture models and regression tree.
We have developed methods and software for the analysis of the data from a state-of-the art magnetic resonance imaging (MRI) technique including MRI, functional MRI, and diffusion tensor image. We have developed and enhanced tools in data mining, Monte Carlo method, statistical modeling, and applied them to scientific problems to understand the function and structure of the brain. Our collaborators and we work closely to study healthy and neurologically disordered children and adults.