Scenario Weights for Importance Measurement (<b>SWIM</b>) – an <tt>R</tt> package for sensitivity analysis

Abstract The Scenario Weights for Importance Measurement ( SWIM ) package implements a flexible sensitivity analysis framework, based primarily on results and tools developed by Pesenti et al . (2019). provides stressed version of stochastic model, subject to model components (random variables) fulfilling given probabilistic constraints (stresses). Possible stresses can be applied moments, probabilities events, risk measures such as Value-At-Risk Expected Shortfall. operates upon single set simulated scenarios from returning scenario weights, which encode the required stress allow monitoring impact all components. weights are calculated minimise relative entropy with respect baseline applied. As well calculating models, including plotting facilities evaluation measures. does not require additional evaluations simulation or explicit knowledge its underlying statistical functional relations; hence, it is suitable black box models. capabilities demonstrated through case study credit portfolio model.