Proposed a new family of generalized linear models for continuous proportional data, named continuous binomial (cobin) regression and its extension, which significantly improves robustness and scalability over traditional beta regression. The implementation is available in the R package cobin; check out install.packages("cobin") in R!
Research Experience
Postdoctoral Associate in the Department of Statistical Science at Duke University, working with Prof. David Dunson.
Education
Ph.D. in Statistics from the Department of Statistics, Texas A&M University, advised by Prof. Huiyan Sang, and collaborated closely with Dr. Eun Sug Park.
Background
My research focuses on developing statistical and machine learning methods that are robust across diverse data conditions, scalable for analyzing large, high-dimensional data with complex dependence structures, and informative for scientific decision-making, with applications to environmental epidemiology and ecology.