D. Ahn
In modern financial systems, stress testing has been considered an important tool to figure out the effect of multiple economic factors on the stability of financial institutions. In usual stress testing, by applying extreme-yet-plausible stress scenarios, we compute risk measures that might not be easily captured by analyzing historical market data or by using stochastic models for market prediction. However, due to the complicated nature of the financial systems, it is hard to identify stress scenarios that cause large losses and threaten the stability of the financial system. Such identification of extreme-yet-plausible scenarios, called reverse stress testing, can help us understand the potential triggers of risky events and remove the arbitrariness in the scenario selection for stress testing. The aim of this project is thus to provide an optimization approach to reverse stress testing, i.e., choosing the most likely scenarios among scenarios that cause a risk measure exceeding a given threshold.