The MCQMC Conference is a bienniel meeting on Monte Carlo and quasi-Monte Carlo methods. It usually attracts 150 to 200 mathematicians, computer scientists, statisticians and researchers in related fields. There is more information in the about MCQMC tab. The conference focusses on the topics below.
CDAR Affiliate Alex Shkolnik will present Important Sampling for Default Timing Models
Abstract: In credit risk management and other applications, one is often interested in estimating the likelihood of large losses in a portfolio of positions exposed to default risk. A serious challenge to the application of importance sampling (IS) methods in this setting is the computation of the transform of the loss on the portfolio. This transform is intractable for many models of interest, limiting the scope of standard IS algorithms. We develop, analyze and numerically test an IS estimator of large-loss probabilities that does not require a transform to be computed. The resulting IS algorithm is implementable for any reduced form model of namy-by-name default timing that is amenable to simulation.