管理科学与工程系学术讲座

题目：Generalized likelihood ratio method and its application in model calibration

主讲人：彭一杰，北京大学Assistant Professor

时间：2018年1月17日 下午4:00-5:00 

地点:同济大厦A楼306室

报告内容摘要

We propose a generalized likelihood ratio (GLR) estimator in a framework that can handle discontinuous sample performances with structural parameters. This work extends the three most popular unbiased stochastic derivative estimators: (1) infinitesimal perturbation analysis (IPA), (2) the likelihood ratio (LR) method, (3) the weak derivative method, to a setting where they did not previously apply. The GLR estimator preserves the single-run efficiency of the classic IPA-LR estimators in applications. Particularly, we apply the GLR estimator to estimate density and its derivatives by Monte Carlo simulation, which in turn leads to a gradient-based simulated maximum likelihood estimation to estimate unknown parameters in a stochastic model without assuming that the likelihoods of the observations are available in closed form. Gradient-based simulated maximum likelihood estimation is flexible in handling various types of model structures.

报告人简介

Dr. Yijie Peng is currently an assistant professor of the Department of Industrial Engineering and Management at Peking University (PKU). He received his Ph.D. from the Department of Management Science at Fudan University and his B.S. degree from the School of Mathematics at Wuhan University. Before joining PKU, he worked as an assistant professor at George Mason University, and postdoctoral scholar at Fudan University and R.H. Smith School of Business at University of Maryland at College Park. Many of his publications appear in high-quality journals including Operations Research, IEEE Transactions on Automatic Control, INFORMS Journal on Computing, Journal of Discrete Event Dynamic System, and Quantitative Finance. His research interests include sensitivity analysis and ranking and selection in the field of simulation optimization, with applications in manufacturing and financial engineering.