The Division of Science and Technology (DST) held a lecture on “A Polynomial Algorithm for Best Subset Selection” with guest speaker Prof. Xueqin Wang on September 22, 2021. Prof. Wang talked about how to choose the best subset, introduced ABESS algorithm and made comparison with other existing methods.

Prof. Wang giving a lecture

Prof. Wang is a professor of the School of Management at the University of Science and Technology of China. He currently serves as a member of the Steering Committee for Statistics Education, Ministry of Education, and an associate editor of the international statistical journals JASA, SII and CJS.

Prof. Wang pointed out that best-subset selection aims to find a small subset of predictors that lead to the most desirable and pre-defined prediction accuracy in a linear regression model. He described best-subset selection with a given support sizesand gave an elaborate introduction to the ABESS algorithm, which helps to determine the size ofs.

Prof. Wang applied the ABESS algorithm to practice and showed the comparison results, and concluded that ABESS algorithm has significant advantages in calculation speed and the number of dimensions. He also discussed the disadvantages of the algorithm.

Prof. Wang telling the comparison results

At the end of the lecture, Prof. Wang interacted with the audience and answered their questions.

Prof. Wang (left) answering Dean Jianxin Pan’s (right) question

Prof. Kaitai Fang (right) presenting UIC souvenir

Prof. Wang and UIC teachers

Reporter: Xinhui Liu

Photographers: Ivy Liao, Xinhui Liu

Editors：Ms. Garbo Hu and Ms. Guoqiu Zhang