Professor of Statistics

Seoul National University

Interested in statistical theory and methods for Non-Euclidean, High-Dimensional data analysis, and Data Privacy

My publications Courses Lab

Brief Biography

I am a Professor of Statistics at the Seoul National University. Before I joined Seoul National University, I spent seven years at the University of Pittsburgh, after completing my PhD at the University of North Carolina at Chapel Hill.

Research interest lies in the theoretical study and applications of modern Statistics and Data Science in the analysis of data that lie on non-standard spaces. This context includes the high-dimension, low-sample-size (HDLSS) situation, non-Euclidean data analysis, the interplay between geometry and statistics, and data fusion. In particular, models and methodologies for dimension reduction, visualization of important variation and hypothesis testings need to be developed with special care for these modern data situations. Particular applications include analysis of directions, landmark-based and skeletally-modeled object shapes, data in stratified spaces or from multiple sources, and retrieving low-dimensional geometric structures in high-dimensional data. I am also interested in statistical issues in Data Privacy, including Differential Privacy and Synthetic Data Generation.

I have authored two books and coauthored one more, all in Korean: book art book art book art

Recent News (More news)

  • 2025-07-29. An article scrutinizing mathematical, computational and statistical aspects of ``Huber means on Riemannian manifolds’’ is now accepted for publication in JRSS-B. Kudos to Jongmin Lee at Pusan National University, my collaborator and the lead author on this!

  • 2025-07-18. The last piece of the endeavors on multiple compositional regressions is now accepted for publication in Annals of the Institute of Statistical Mathematics. This work is on the group-wise significance tests utilizing a newly-developed grouped-debiasing procedure for multiple compositional regression models, and was a part of Sujin’s dissertation.

  • 2025-06-29. A research paper on adaptive reference-guided estimation of high-dimensional PC subspace is accepted for publication in Stat. This work constitutes Dongsun’s masters thesis.

  • Last updated: August 06, 2025