Online Program Home
My Program

Abstract Details

Activity Number: 156 - Statistical Aspects in Stochastic and Deterministic Simulation
Type: Topic Contributed
Date/Time: Monday, July 30, 2018 : 10:30 AM to 12:20 PM
Sponsor: Section on Physical and Engineering Sciences
Abstract #330219
Title: Using Concomitant and Nested Simulation for Tail Risk Measure Estimation
Author(s): Mingbin Feng*
Companies: University of Waterloo
Keywords: Stochastic Simulation; Monte Carlo; Experiment Design; Variable Annuities; Machine Learning; Clustering
Abstract:

Tail risk measures is of critical importance for enterprise risk management, especially for managing large portfolios of complex financial instruments. The computational burdens required by such simulation can be substantial or even unbearable, depending on the complexity of the underlying economic model and the risk management objective. This paper proposes, analyzes, and tests an efficient nested simulation procedure for estimating tail risk measures. The procedure first uses proxy models and their concomitants to quickly and accurately identify tail scenarios where the given computational budget is concentrated. We demonstrate the proposed procedure in estimating tail risk measures of variable annuities. Our results show that, given a fixed computational budget, the proposed procedure can be an order of magnitude more accurate that a standard nested simulation procedure.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program