1. 2020+ Thien M. Le and PingShou Zhong Goodness of Fit Tests for Network Structures. Preprint.
2. 2020+ Thien M. Le and PingShou Zhong Estimation of Precision Matrix with Known Graphical Structure. Preprint.
3. 2020+ F. Claire Hankenson, Joshua Kim, Thien M. Le, Frank Lawrence, and Jacquelyn M. DelValle Guidance on Skin Preparation
Methods and Thermal Support for Surgery in Laboratory Mice and Rats. In Revision.
4. 2020+ Hyokyoung G. Hong, Wodan Ling, Thien M. Le, Hong Su An, Kathleen Oberst, and Ann M Annis "Quantile regression
exploration of heterogeneous associations between sociodemographics, multimorbidities, and functional limitations among
older adults, NHANES, 20072016" . In Revision.
PUBLICATIONS
RESEARCH INTERESTS

Network Science

HighDimensional Statistical Inference

Approximate Bayesian Computation

Statistics with Applications in Healthcare, Finance, and Agriculture
Many realworld systems are composed of components link in some way. For example, a collection of computers linked by data connections or people linked by an acquaintance. Understand the pattern of connection is essential to understand how the systems work.
My current research focuses on developing statistical tools to understand these connecting patterns, and how to apply these patterns of connection to improve the precision in statistical modeling and decision making.
Another focus of my research is highdimensional statistical inference. Highdimensional data sets are data sets with the number of variables bigger than the number of observations. For instance, in biology, genetics data sets often contain thousands of genes as variables, but there are only several hundreds of patients as observations. Or in finance, scanner data sets per household have a large number of products purchased as variables, but only a couple of purchases per month as observations. Traditional statistics were not designed to cope with this type of data, which brought my attention to the field. My Ph.D. dissertation studied the problem of estimating and testing for highdimensional Gaussian graphical models. In my work, my Advisor and I developed the first global test to test a prespecified Gaussian graphical model for highdimensional data. Currently, I am also interested in studying the testing problem for different populations and the change point detection problem of the dynamic Gaussian graphical models.
Besides, I am also interested in approximate Bayesian computation and statistics with applications in healthcare, finance, and agriculture.