Stable and Monotonic Allocation in Risk Pooling Problems
In many practical scenarios, agents cooperate to enjoy the benefits of risk pooling. Two notable examples are inventory pooling of newsvendors and the risk sharing in, say, reinsurance market. A successful cooperation critically depends on how one allocates the costs among the cooperating agents. One consideration is the stability issue, i.e., no agent nor a coalition of agents should be better off by operating on its own. Another consideration is the monotonicity issue, i.e., as a new agent joins the coalition, every member in the coalition should be better off. In this talk, I will discuss these two issues in the cooperation of newsvendors as well as the risk sharing problem.
Zhenyu Hu received his Ph.D. in Industrial Engineering from the University of Illinois at Urbana-Champaign (UIUC). Prior to that, he obtained his B.Sc. in Applied Mathematics from Sun Yat-sen University. He worked as a Research Summer Intern at IBM T.J. Watson Research Center in 2014. Zhenyu has worked on the dynamic pricing problems with reference price models, and recently he is also interested in cooperative game theory and its application in supply chain management.