The investigator and his colleagues will study selected topics in high-dimensional regression, classification and hypothesis testing. Specifically, these include a sparse factor model for high-dimensional regression and survival analysis, large scale interaction testing with applications to genomics, the computational algorithms for the fused lasso, and a research monograph on L1-constrained (Lasso) methods in statistics.
This work will help scientists working in biotech and other areas, who generate large scale datasets, to interpret and uncover the important patterns in their data. This should help scientists and doctors to discover the biological bases of many diseases, and improve prognosis and treatment selection for patients.
