I develop statistical approaches to make inference from large-scale, complex data sets. For example, how can we infer model coefficients when the number of variables is large or even comparable to the number of observations? How can we identify genetic variants associated with severe mental disorders? How can we provide disease risk prediction are equally accurate across different populations? Statistical methods developed for these questions be used not only in biomedical sciences, but also in the context of social sciences and policy decision making. In additional to research, I am passionate about data science education, and working with students to tackle data science projects with positive social impact. Previously, I have worked with journalists to identify behavioral health conditions in police records, and predicting platelet blood bag demand in hospitals.
