Comparing Standard and Extreme VaR Models During Highly Volatile Periods

A bank’s Value at Risk (VaR) determines the amount of capital that it must set aside to satisfy bank regulations. VaR relates to the size of potential trading losses; the greater the VaR, the greater a trading loss the bank could suffer and consequently the more capital it must set aside. This reduces the probability of default, but it also reduces the potential scope of derivative trading operations, reducing potential profits. VaR is obtained from a mathematical model for the joint evolution of all risk factors influencing the value of the bank’s trading portfolio. Unfortunately, in periods of market stress of the type experienced recently, VaR numbers produced by conventional models are unreliable. However, an alternative modeling approach based on Extreme Value Theory (EVT) has recently shown promise. This project will research the practical application of this methodology to a range of trading portfolios. The project will evaluate the performance…TOBECONT’D

Faculty Supervisor:

Andrey Pavlov

Student:

Partner:

Markit

Discipline:

Business

Sector:

University:

Simon Fraser University

Program:

Accelerate