A Benchmark Model for Market Risk

Christian Meyer and Peter Quell

Empirical data indicates that financial markets are not stationary. For QRMs to be useful in the context of market risk, one has to take this aspect into account. At the end of the last chapter, the filtered historical simulation was discussed to address this aspect of non-stationarity. Another potential solution to this problem is the revised RiskMetrics model (Zumbach, 2007). In a certain way, both models capture time-varying characteristics of markets by starting with the dynamics of risk factors. In this sense, they provide a bottom-up perspective.

In this chapter, a benchmark model for market risk will be presented that is inherently of a top-down flavour. The starting point will be the time series of profits and losses on a portfolio level, a dataset that every bank running an internal model already has in stock. That implies several advantages:

• the benchmark model is easily implemented with existing internal models;

• since the benchmark model provides a top-down approach, it is a useful companion to other risk model validation processes; and

• the benchmark model can focus on the distributional properties of the portfolio P&L in contrast to the distributional