Journal of Risk

The papers contained in this issue of The Journal of Risk address the trade-offs involved in model complexity. The first two do so within a context that accounts for capital requirements, while the latter two focus on pricing and hedging accuracy.

In the first paper in this issue, "The relationship between credit default swap spreads and equity prices", Michele Marzano, Gary Dunn and Nick Constantinou conduct an empirical study that illustrates the impact of the valuation of credit default swaps (CDSs) on regulatory capital. The authors show that, for major markets, their econometric specification leads to stable estimates of the sensitivity of the changes in the CDS spreads relative to a CDS index and the changes in corresponding share prices relative to general market indexes. They also show that their model generates market consistent credit-adjusted values-at-risk.

Financial institutions generally have discretion in their choice of the rating technology used to evaluate the credit risk of borrowers. It is reasonable, then, to assume that their decisions reflect the trade-off between a potential competitive advantage due to better technology and its associated high cost of implementation. In our second paper, "Choice of rating technology and loan pricing in imperfect credit markets", Hannelore De Silva, Engelbert Dockner, Rainer Jankowitsch, Stefan Pichler and Klaus Ritzberger use a two-stage game model in an oligopolistic banking sector to evaluate this trade-off. The authors show that better technology does not necessarily translate into increased profits and that identical banks might adopt such technology sequentially, rather than simultaneously. They also show that increased competition in the banking sector may in fact impede the adoption of better rating technology.

The third paper in this issue, "Are traders'rules useful for pricing options? Evidence from intraday data" by Sol Kim, is focused on comparing the pricing accuracy of ad hoc models used by traders against that of more complex models. The author shows that, for intraday data, traders who use the standard Black-Scholes values adjusted heuristically significantly misprice very liquid options relative to models that account for stochastic volatility and jumps.

In the fourth and final paper of the issue, "Numerical experiments on hedging cliquet options", Fiodar Kilin, Morten Nalholm and Uwe Wystup conduct a simulation study to compare the pricing and hedging performance of various models used to evaluate cliquet options. The latter are significantly affected by forward volatility and skew. The results obtained in this study indicate that the standard models for stochastic volatility - such as those of Heston and of Barndorff-Nielsen and Shephard, as well as the variance gamma specification - perform in a manner that can be severely detrimental, thus underlining the genuine need for alternative approaches.

Farid AitSahlia
Warrington College of Business Administration,
University of Florida