My primary area of research involves the development of novel Bayesian statistical methods for analyzing complex data structures.  Such data structures of interest include longitudinal data, missing data, correlated data, clustered data, and high-dimensional data.  In particular, my work involves utilizing modeling choices with prior structures designed to provide computational stability and/or interpretable model results, including the use of variable selection and variable shrinkage models.  A critical component of this work is the use and development of computationally efficient algorithms that can be used to perform inference in these cases, such as Markov chain Monte Carlo sampling.  

In addition to my methodological work, I have active research collaboration across the university, particularly within the School of Medicine.  At current, I have collaborative agreements with the Depts of Surgery; Obstetrics, Gynecology, and Women’s Health; and Radiation Oncology.  I have also been involved in projects with psychology, pathology, and other groups.