Dr. Lin's major statistical research interests lie in developing statistical methods for high-dimensional and correlated data. Examples of high-dimensional data include genomic and proteomic data in basic, population and clinical sciences. Examples of correlated data include longitudinal data, clustered data, hierarchical data and spatial data. She is particularly interested in developing statistical and computational methods for "omics" data in population-based studies, such as genetic epidemiology, genetic environmental sciences and clinical studies. She currently serves as the coordinating director of the Program of Quantitative Genomics of Harvard School of Public Heath http://www.hsph.harvard.edu/pqg. Her statistical research is funded by the MERIT award from the National Cancer Institute on "Statistical Methods for Correlated and High-Dimensional Biomedical Data" http://www.cancer.gov/researchandfunding/MERIT/Lin
Dr. Lin's specific areas of statistical research include statistical learning methods for high-dimensional data, dimension reduction, variable selection, nonparametric and semiparametric regression models, measurement error, mixed (frailty) models, estimating equations, and missing data.
Dr. Lin's areas of applications include cancer, genetic epidemiology, gene and environment, genome-wide association studies, genomics in population science, biomarkers and proteomics.