I study how complex data can be described by models whose components have a clear statistical meaning. My research develops methods for selecting nonlinear covariate effects, understanding associations across a response distribution, and estimating shared structure in high-dimensional data.
These questions connect nonparametric statistics, model selection, and statistical learning. I develop estimation procedures and computational algorithms, and study the conditions under which their conclusions are reliable.
Professor of Statistics
Chair, Department of Statistics
Contact
erlee@skku.edu
Room 32412, Dasan Hall of Economics
Seoul, Korea
University research profile
Academic biography
Which variables matter when their effects are unknown functions? I study component selection, smooth backfitting, and uncertainty quantification for flexible regression.
Annals of Statistics · 2024
Efficient functional lasso kernel smoothing for high-dimensional additive regression
Which associations change across a response distribution? I investigate quantile regression, model selection, and varying coefficient models.
Journal of the American Statistical Association · 2014
Model selection via Bayesian information criterion for quantile regression models
How can many binary or count responses be analyzed together? I develop structured models that reduce dimension while retaining relevant covariates and response patterns.
Biometrika · 2021
Poisson reduced-rank models with an application to political text data
Journal of the American Statistical Association · 2024
Low-rank regression models for multiple binary responses and their applications to cancer cell-line encyclopedia data
Complete publication list · Students and alumni · Teaching