Computation of Risk Measures via Efficient Least-squares Monte Carlo

The computation of risk profiles for financial products and portfolios is an extremely important problem, both for regulatory and internal management purposes. For complex products whose value depends on a number of underlying risk factors and for which exercise decisions can be made prior to maturity, Monte Carlo simulation techniques are the only viable procedures. This project aims to adopt a simulation method used for pricing products, to computing risk exposures. Various ways of improving the computational speed will be explored. Having a highly-trained intern, with particular expertise in this area, allows for our industrial partner to take on this important longer-term project that would otherwise not be possible due to resource constraints. Completion of this project will significantly improve the ability of the risk management team to assess and manage the risks of complex products.

Felix Kan
Faculty Supervisor: 
Dr. Mark Reesor