Hostname: page-component-7c8c6479df-995ml Total loading time: 0 Render date: 2024-03-28T16:21:36.078Z Has data issue: false hasContentIssue false

Scenario Weights for Importance Measurement (SWIM) – an R package for sensitivity analysis

Published online by Cambridge University Press:  12 May 2021

Silvana M. Pesenti*
Affiliation:
Department of Statistical Sciences, University of Toronto, Toronto, M5G 1X6, Canada
Alberto Bettini
Affiliation:
Assicurazioni Generali S.p.A, Trieste, 34132, Italy
Pietro Millossovich
Affiliation:
DEAMS, University of Trieste, Trieste, 34127, Italy Faculty of Actuarial Science and Insurance, The Business School (formerly Cass), City, University of London, London, EC1Y 8TZ, United Kingdom
Andreas Tsanakas
Affiliation:
Faculty of Actuarial Science and Insurance, The Business School (formerly Cass), City, University of London, London, EC1Y 8TZ, United Kingdom
*
*Corresponding author. E-mail: silvana.pesenti@utoronto.ca

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

Type
Paper
Copyright
© The Author(s), 2021. Published by Cambridge University Press on behalf of Institute and Faculty of Actuaries

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

Beckman, R.J. & McKay, M.D. (1987). Monte Carlo estimation under different distributions using the same simulation. Technometrics, 29(2), 153160.CrossRefGoogle Scholar
Borgonovo, E., Hazen, G.B. & Plischke, E. (2016). A common rationale for global sensitivity measures and their estimation. Risk Analysis, 36(10), 18711895.CrossRefGoogle ScholarPubMed
Borgonovo, E. & Plischke, E. (2016). Sensitivity analysis: a review of recent advances. European Journal of Operational Research, 248(3), 869887.CrossRefGoogle Scholar
Breuer, T. & Csiszár, I. (2013). Systematic stress tests with entropic plausibility constraints. Journal of Banking & Finance, 37(5), 15521559.CrossRefGoogle Scholar
Cambou, M. & Filipović, D. (2017). Model uncertainty and scenario aggregation. Mathematical Finance, 27(2), 534567.CrossRefGoogle Scholar
Csiszár, I. (1975). I-divergence geometry of probability distributions and minimization problems. The Annals of Probability, 3(1), 146158.CrossRefGoogle Scholar
Emmer, S., Kratz, M. & Tasche, D. (2015). What is the best risk measure in practice? A comparison of standard measures. Journal of Risk, 18(2), 3160.CrossRefGoogle Scholar
Iooss, B., Janon, A. & Pujol, G. (2019). Sensitivity: global sensitivity analysis of model outputs. R package version 1.16.2.Google Scholar
McNeil, A.J., Frey, R. & Embrechts, P. (2015). Quantitative Risk Management: Concepts, Techniques and Tools-revised edition. Princeton, New Jersey: Princeton University Press.Google Scholar
Pesenti, S.M., Bettini, A., Millossovich, P. & Tsanakas, A. (2020). SWIM: Scenario Weights for Importance Measurement. R package version 0.2.0.Google Scholar
Pesenti, S.M., Millossovich, P. & Tsanakas, A. (2016). Robustness regions for measures of risk aggregation. Dependence Modeling, 4(1), 348367.CrossRefGoogle Scholar
Pesenti, S.M., Millossovich, P. & Tsanakas, A. (2019). Reverse sensitivity testing: what does it take to break the model? European Journal of Operational Research, 274(2), 654670.CrossRefGoogle Scholar
R Core Team (2019). R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria.Google Scholar
Saltelli, A., Ratto, M., Andres, T., Campolongo, F., Cariboni, J., Gatelli, D., Saisana, M. & Tarantola, S. (2008). Global Sensitivity Analysis: The Primer. West Sussex, England: John Wiley & Sons.Google Scholar
Tsanakas, A. & Millossovich, P. (2016). Sensitivity analysis using risk measures. Risk Analysis, 36(1), 3048.CrossRefGoogle ScholarPubMed
Vallender, S. (1974). Calculation of the wasserstein distance between probability distributions on the line. Theory of Probability & Its Applications, 18(4), 784786.CrossRefGoogle Scholar
Villani, C. (2008). Optimal Transport: Old and New, vol. 338. Berlin Heidelberg: Springer Science & Business Media.Google Scholar
Weber, S. (2007). Distribution-invariant risk measures, entropy, and large deviations. Journal of Applied Probability, 44(1), 1640.CrossRefGoogle Scholar
Xie, Y. (2019). bookdown: Authoring Books and Technical Documents with R Markdown. R package version 0.12.Google Scholar