Updates and Information on Coronavirus (COVID-19)

Faculty Profile: Wei Zheng

March 20, 2018

When Wei Zheng joined the Haslam College of Business faculty in 2017, he brought a wealth of mathematical and statistical research experience with him.

“My career path is not typical in a business college,” Zheng says. “I was a statistics major in a math department for both my undergraduate and graduate studies, and my first job was in that environment.”

Zheng’s early research centers on optimal experiment design with applications typically in clinical trials.

“The main purpose of experimental design is to collect the data in the most cost-effective and informative way,” explains Zheng, who published the research in Annals of Statistics and Journal of the American Statistical Association. “Our approach is to formulate the design problem into an optimization problem based on some statistically meaningful criteria. For a type of design problems, I established a unified way of deriving optimal or efficient designs under various scenarios.”

Recently, he is also extending his research interest to the interface between design and computing in two ways. The obvious one is to develop fast algorithms to derive optimal or efficient designs. The other is the opposite: use the design idea to improve on computational performance of the methods in both statistics and machine learning.

Joining a business faculty lines up with the recent shift of Zheng’s research interests.

“My approach used to be very theoretical,” he says. “Now, I care more about the real applications and impact of my research. My research could be useful in a lot of new problems from the business world.”

Zheng currently teaches the Master of Science in Business Analytics capstone course along with undergraduate and graduate level courses on probability and experiment design. He enjoys sharing his expertise as students design their own experiments in class projects.

“We start with the assumption that we must be able to leverage the results of the experiment for decision making,” he says. “That knowledge leads us to ask, methodologically speaking, how can we make the best designed experiment in the beginning?”