In stress testing financial models for extreme disaster conditions, what advanced statistical technique should be used to derive plausible but extreme input parameters for variables like sales decline and cost increases?

The advanced statistical technique that should be used to derive plausible but extreme input parameters for variables like sales decline and cost increases in stress testing financial models for extreme disaster conditions is Extreme Value Theory. Extreme Value Theory, or EVT, is a specialized branch of statistics specifically designed to model the behavior of rare events, which occur in the extreme tails of a probability distribution. Unlike traditional statistical methods that often focus on the central part of data and assume a normal or similar distribution, EVT provides a rigorous framework for understanding and quantifying the likelihood and magnitude of exceptionally large or small observations. This makes it uniquely suited for determining "plausible but extreme" values, as it focuses precisely on those rare, high-impact events that are central to extreme disaster scenarios.