1. 주 제 : Dynamic Importance Sampling for Affine Point Processes

2. 연 사 : 라무중박사 (Barclays)

3. 내 용 : Stochastic models of event timing are common to many areas, including finance, insurance, queuing and reliability. We  adopt a marked point process setting to model the arrival of events in time. In these models, the arrival of an event is governed by an intensity or conditional event arrival rate. The intensity is a modeling primitive. A constant intensity, for example, generates a Poisson model. A stochastic intensity process generates richer event behavior. Every event is associated with random variable, called a mark, that reveals further, event-specific information. We develop efficient importance sampling estimators of certain rare-event probabilities involving affine point processes. Our approach is based on dynamic importance sampling, and the design of the estimators departs from past literature to accommodate the point process setting. Namely, the state-dependent change of measure is updated not at event arrivals but over a deterministic time grid. Several common criteria for the optimality of the estimators are analyzed. Numerical results illustrate the advantages of the proposed estimators in an application setting.

4. 일 시 : 2017년 09월 08일 (금) 오후 2:00

5. 장 소 :

6. 문 의 : 장우진교수 (changw@snu.ac.kr)