Journal of Risk

Welcome to Volume 9 Issue 3 of The Journal of Risk. This issue is made up of 4 technical papers: There are now a few parameters to capture risk and related quantities, such as risk premium. Estimating them is an on-going topic of investigation. All four papers in this issue deal with empirical evaluations of risk parameters using different approaches.

The first paper, by James S. Doran, develops a delta-gamma-based approach to estimate volatility risk premium. A major advantage provided by this method is its incorporation of gamma, which reduces discretization error. Thus, with more focus on specification error, a sharper risk premium is obtainable.

The second paper, by Chulwoo Han, Frank C. Park and Jangkoo Kang, discusses the delta-gamma-based method to quickly evaluate VAR for mortgagebacked securities. This method is shown to be significantly more efficient than a full-blown simulation approach. Another contribution provided by the authors is an actual implementation that illustrates the interplay between models, such as cashflow and interest rates, and the importance of their mutual consistency.

Interest rate risk exposure is the topic addressed by the next authors, Roberta Fiori and Simonetta Iannotti. Their paper applies principal component analysis in a Monte Carlo context to estimate VAR. In so doing, they provide evidence supporting the superiority of a non-parametric approach to satisfy consistency with Basel II guidelines.

In the last paper, Ebenezer Asem investigates the impact of inefficient parameter estimates on standard deviation forecasts and, ultimately, on VAR forecasts. Given the well-known effects of a misspecified Gaussian likelihood function on GARCH parameters, the goal of the paper is to assess the ultimate impact of this misspecification on VAR. The author uses Monte Carlo simulation and equity data to support the conclusion that efficient VAR estimates still result.