An Integrated Stress Testing Framework Via Markov Switching Simulation

The capturing of tail events, especially those that incur severe loss at rare chance, is one of the important objectives for modern risk analysis. However past behavior in financial data is not necessarily a correct reflection of the possible scenarios in the future. The economic turmoils in the recent years have called for a more forward looking approach to financial risk management that integrates expert knowledge on plausible future scenarios with classical risk management models calibrated on past behavior. As a complementary risk analysis tool, stress testing is getting more and more attention from both regulators and practioners. Nevertheless, classical risk analysis models, such as VaR models, based on historical data and stress testing are often disconnected. This disconnection can prevent a comprehensive view of the risk profile of a financial institution. This paper proposes a multi-period switching simulation based method for integrated stress testing risk analysis that incorporate plausible events that are not necessarily captured in history or in historical stressed calibration of risk models. An integrated risk model and stress testing framework not only leads to forward-looking tail risk measurement that mitigates the "Black Swan" effect, but also takes stress testing into advanced risk management decision making analysis like scenario based portfolio optimization