Journal of Risk Model Validation

The December issue of The Journal of Risk Model Validation consists of one backtesting paper and three papers on value-at-risk. There are a number of subthemes that involve at least two of the papers: namely, risk measurement in frontier and Asian stock markets and the use of stochastic volatility models (especially the broad family known as generalized autoregressive conditional heteroscedasticity (GARCH) models). These models are often derided by practitioners, and it is true that they do not work uniformly well in all situations. However, in my view they are a useful addition to standard risk tools, especially in relatively short-term situations.

I shall discuss the backtesting paper first. "Backtesting for counterparty credit risk" by Sebastian Schnitzler, Niklas Rother, Holger Plank and Peter Glößner covers a wide range of issues in backtesting with a particular focus on long-term exposure validation and issues to do with long time horizons. This is an area in which there are many unresolved methodological issues and where it is clear that data availability and the careful treatment of existing data are vital if reasonable results are to be obtained. The authors make a valuable contribution to the literature.

The issue's second paper, and the first of the value-at-risk ones, is by Sunny B. Walter Madoroba and Jan W. Kruger and is titled "Liquidity effects on value-at-risk limits: construction of a new VaR model". As the paper's title suggests, the authors provide a new VaR model that incorporates intraday price movements on high-low spreads as well as other adjustments. They apply their model to the Johannesburg stock exchange. The model is tested using some interesting statistical approaches and is claimed to be both valid and robust.

The third paper in the issue, "Forecasting value-at-risk for frontier stock market indexes using GARCH-type models and extreme value theory: model validation for dynamic models" by Dany Allan Nicholas Ng CheongVee, Preethee Nunkoo Gonpot and Noor Ul Hacq Sookia involves analysis that uses rival models of stochastic volatility as well as extreme value theory to capture fat tails. The data used is returns from frontier stock market indices. This is interesting in its own right as it is a new area of research generating new models and new techniques. The benefit of such exercisesmis that they shed light on existing markets by providing information on parameter regions, which are not normally explored by conventional models.

Our final paper, "The interrelation of stock markets in China, Taiwan and Hong Kong and their constructional portfolio's value-at-risk estimate" by Jung-Bin Su, to some extent complements the previous paper. Again it considers stochastic volatility models (in this case GARCH) but focuses on correlations between the three markets in the title of the paper and suggests appropriate adjustments to account for spillovers, feedback effects and other such issues that are likely to arise in this situation.

Steve Satchell
Trinity College, University of Cambridge

You need to sign in to use this feature. If you don’t have a account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an individual account here