PhD (Statistics), University of North Carolina at Chapel Hill, Chapel Hil, 2011
It is well known that modern statistical problems are increasingly high dimensional, i.e. the number of parameters is large. Analyses of high dimensional data raise many methodological issues not present, or not as important, in lower dimensional settings. Much of my research focuses on developing statistical machine learning methods to analyze high dimensional data. I am also interested in developing computationally efficient statistical methods to solve dynamical models in infectious disease study. I have developed efficient Bayesian frameworks to estimate parameters in ordinary differential equations and novel computational approaches to pinpoint predisposed recombination regions in HIV. Moreover, I have been seeking every opportunity to collaborate with scientists from a wide variety of fields on using statistical methods to solve real problem.