ABSTRACT
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 356
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 356
S TRESS TESTING IS A TERM used in financial practice without any generally accepteddefinition. It appears in the context of quantification of losses or risks that may appear under special, mostly extremal circumstances (Kupiec 2002). Such circumstances
are described by certain scenarios which may come from historical experience (a crisis
observed in the past)—historical stress test, or may be judged to be possible in the future
given changes of macroeconomic, socioeconomic or political factors-prospective stress
test, etc. The performance of the obtained optimal decision is then evaluated along these,
possibly dynamic, scenarios or the model is solved with an alternative input. Stress testing
approaches differ among institutions and also due to the nature of the tested problem and
the way in which the stress scenarios have been selected. In this chapter, we focus on the
stress testing of two risk measures, VaR and CVaR, giving the ‘test’ a more precise
meaning. This is made possible by the exploitation of parametric sensitivity results and
the contamination technique.
