Southwest Jiaotong University School of Mathematics


首页  >  学术科研  >  学术交流  >  正文

Calibration of Distributionally Robust Empirical Optimization Models

来源:   作者:     日期:2019-08-30 15:49:59   点击数:  

报告题目:Calibration of Distributionally Robust Empirical Optimization Models

报告人:Prof. Jun-ya Gotoh (Chuo University, Japan) 後藤順哉 教授 (中央大学)






We study the out-of-sample properties of robust empirical optimization and develop a theory for data-driven calibration of the “robustness parameter" for worst-case maximization problems with concave reward functions. We show that the first-order benefit of “little bit of robustness" is a significant reduction in the variance of the out-of-sample reward while the corresponding impact on the mean is almost an order of magnitude smaller. One implication is that a substantial reduction in the variance of the out-of-sample reward is possible at little cost if the robustness parameter is properly calibrated. To this end, we introduce the notion of a robust mean-variance frontier to select the robustness parameter and show that it can be approximated using resampling methods like the bootstrap. Our examples show that robust solutions corresponding to the ambiguity parameter that optimizes an estimate of the out-of-sample expected reward with no regard for the variance are often insufficiently robust.

This is a joint work with Michael Jong Kim and Andrew E. B. Lim.


 Jun-ya Gotoh received his B. Eng., M. Eng., and PhD in Industrial Engineering from Tokyo Institute of Technology in 1996, 1998, and 2001, respectively. He was with the University of Tsukuba, Japan, from 2001 to 2007, where he worked as an assistant professor. He has been with the Industrial and Systems Engineering Department at the Chuo University since 2007-present as an associate professor (2007-2014) and a full professor (2014 to present). His research interests lie in optimization modelling for decision making under uncertainty, data analytics and financial engineering.

Dr. Gotoh is a recipient of the 7th Research Award (2017) of the Operations Research Society of Japan (ORSJ). He is a Fellow of ORSJ since 2018. His professional activities include: Editorial Board Member of Annals of Operations Research (2015-present); Associate Editor of Optimization and Engineering (2014-present); Associate Editor of the Journal of Operations Research of Japan (since 2008 to 2012). A co-guest editor of a special issue of Annals of Operations Research (Vol.262, No.1, 2018) entitled “Special Volume on Risk Management Approaches in Engineering Applications” and a special issue of the Journal of Operations Research of Japan (Vol.54 No.4, 2011). He worked as a regular member of ORSJ, Japan Financial Economics and Engineering, and INFORMS. Also, he serves as a Local Organizing Committee Member of 5th International Conference on Continuous Optimization (ICCOPT 2016 Tokyo) in Tokyo, August 6-11, 2016.