Research

My methodological and theoretical research focuses on high-dimensional modeling and scalable machine learning algorithms for large and complex data sets. My recent work has focused on developing flexible methods that relax assumptions such as linearity, Gaussianity, and/or independence. Some of the statistical tools I use include nonconvex optimization, sparsity-inducing priors, and decision tree ensembles. My applied research focuses on meta-analysis, spatiotemporal modeling, and analysis of electronic health records.