The tail dependograph
- 47 Downloads
All characterizations of non-degenerate multivariate tail dependence structures are both functional and infinite-dimensional. Taking advantage of the Hoeffding–Sobol decomposition, we derive new indices to measure and summarize the strength of dependence in a multivariate extreme value analysis. The tail superset importance coefficients provide a pairwise ordering of the asymptotic dependence structure. We then define the tail dependograph, which visually ranks the extremal dependence between the components of the random vector of interest. For the purpose of inference, a rank-based statistic is derived and its asymptotic behavior is stated. These new concepts are illustrated with both theoretical models and real data, showing that our methodology performs well in practice.
KeywordsGlobal sensitivity analysis Hoeffding–Sobol decomposition Multivariate extreme value analysis Pairwise index Tail dependency graph
AMS 2000 Subject ClassificationsPrimary 62J10 62G32 62H20 Secondary 62J15 62G20 62-09
Unable to display preview. Download preview PDF.
The authors would like to thank Christian Genest (McGill University, Montréal, Canada) for fruitful discussions. The first author would also like to thank Roland Denis and Benoît Fabrèges (Institut Camille Jordan, Université de Lyon, France) whose discussions and workshop led to great progress in the codes associated with this paper. The authors would like to thank the editor and referees for their helpful comments.
- Browne, T., Fort, J.-C., Iooss, B., Le Gratiet, L.: Estimate of quantile-oriented sensitivity indices working paper or preprint (2017)Google Scholar
- de Haan, L., Ferreira, A.: Extreme value theory. An introduction. Springer series in operations research and financial engineering. Springer, New York (2006)Google Scholar
- Fougères, A.-L.: Multivariate extremes. In: Finkenstädt, B., Rootzén, H. (eds.) Extreme Values in Finance, Telecommunications, and the Environment. Monographs on Stat. and Appl. Prob. 99, Chapter 7, pp. 373-388. Chapman and Hall/CRC (2004)Google Scholar
- Hofert, M., Hornik, K.: qrmdata: data sets for quantitative risk management practice. R package version 2016-01-03-1 (2016)Google Scholar
- Hooker, G.: Discovering additive structure in black box functions. In: Proceedings of KDD 2004, pp. 575–580. ACM DL (2004)Google Scholar
- Huang, X.: Statistics of bivariate extremes. PhD Thesis, Erasmus University Rotterdam, Tinbergen Institute Research series No. 22 (1992)Google Scholar
- Kucherenko, S., Song, S.: Quantile based global sensitivity measures. ArXiv e-prints (2016)Google Scholar
- Pujol, G., Iooss, B., Janon, A.: Sensitivity: global sensitivity analysis of model outputs. R package version 1.14.0 (2017)Google Scholar
- Smith, R.L.: Max-stable processes and spatial extremes. Dept. of Math., Univ. of Surrey, Guildford GU2 5XH England (1990)Google Scholar
- Stephenson, A.G.: Evd: extreme value distributions. R News 2(2), 31–32 (2002)Google Scholar
- Sueur, R., Iooss, B., Delage, Th.: Sensitivity analysis using perturbed-law based indices for quantiles and application to an industrial case working paper or preprint (2017)Google Scholar
- Tiago de Oliveira, J.: Structure theory of bivariate extremes, extensions. Estudos de Matematica, Estatistica, e Economicos 7, 165–195 (1962/63)Google Scholar
- van der Vaart, A.W.: Asymptotic Statistics, volume 3 of Cambridge Series in Statistical and Probabilistic Mathematics. Cambridge University Press, Cambridge (1998)Google Scholar