Statistics faculty member has been noticed as a

Rising Star

By Melody Galen

"I've never seen anything like this," statistics department Chair Nancy Flournoy says of Assistant Professor Min Yang. The accolades from other statisticians of Yang's work literally filled a recommendation letter that Flournoy wrote recently. She says he's modest about his accomplishments, yet she's never seen a better grant record than his at this academic stage.

Yang was awarded his first National Science Foundation grant in 2003, then another in 2006, and he was a winner of the NSF CAREER Award in 2008. The CAREER award is NSF's most prestigious award in support of junior faculty who exemplify the role of teacher-scholars. Amstat News, the membership magazine of American Statistical Association, cited Yang in the September 2008 issue as an up-and-comer, with five other junior statisticians from prestigious institutions all over the country.

Why All the Fuss?

Yang's area of research is optimal design of experiments. This subfield of statistics involves planning experiments carefully to maximize the information that can be gained. Sounds like a no-brainer, right? Who wouldn't want to do that? Yang's research helps others refine their studies to reduce their sample sizes and still get accurate and robust results. For example, experiments such as clinical drug trials are often expensive. Phase III trials can easily involve thousands of patients. The resources required to maintain so many patients could be staggering in terms of time, money and man power. Optimal experimental design protocols not only help reduce the cost by limiting the number of patients needed but also increase the accuracy of these experiments.

Design of experiments is a discipline that has very broad application across all the natural and social sciences, including medicine, biology, agriculture and engineering. Yang's use of cross-over designs and nonlinear mixed-effects models helps ensure that experiments are designed using the optimal number of people and resources and that the study runs for the optimal amount of time.

There is no end to the number of research questions out there. If, in these shaky economic times, someone can divine a way to make experiments and studies more cost effective, who can argue with that?

Links:

Min Yang
Department of Statistics