Abstract
When estimating the risk of a P&L from historical data or Monte Carlo simulation, the robustness of the estimate is important. We argue here that Hampel’s classical notion of qualitative robustness is not suitable for risk measurement, and we propose and analyze a refined notion of robustness that applies to tail-dependent law-invariant convex risk measures on Orlicz spaces. This concept captures the tradeoff between robustness and sensitivity and can be quantified by an index of qualitative robustness. By means of this index, we can compare various risk measures, such as distortion risk measures, in regard to their degree of robustness. Our analysis also yields results of independent interest such as continuity properties and consistency of estimators for risk measures, or a Skorohod representation theorem for ψ-weak convergence.
Similar content being viewed by others
References
Acerbi, C., Tasche, D.: On the coherence of expected shortfall. J. Bank. Finance 26, 1487–1503 (2002)
Belomestny, D., Krätschmer, V.: Central limit theorems for law-invariant coherent risk measures. J. Appl. Probab. 49, 1–21 (2012)
Beutner, E., Zähle, H.: A modified functional delta method and its application to the estimation of risk functionals. J. Multivar. Anal. 101, 2452–2463 (2010)
Biagini, S., Frittelli, M.: A unified framework for utility maximization problems: an Orlicz space approach. Ann. Appl. Probab. 18, 929–966 (2008)
Biagini, S., Frittelli, M.: On the extension of the Namioka–Klee theorem and on the Fatou property for risk measures. In: Delbaen, F., et al. (eds.) Optimality and Risk: Modern Trends in Mathematical Finance. The Kabanov Festschrift, pp. 1–29. Springer, Berlin (2009)
Boussama, F.: Ergodicity, mixing and estimation in GARCH models. Ph.D. thesis, University of Paris 7 (1998)
Breiman, L.: Probability. Classics in Applied Mathematics, vol. 7. SIAM, Philadelphia (1991) (corrected reprint of the 1968 original)
Cheridito, P., Li, T.: Risk measures on Orlicz hearts. Math. Finance 19, 189–214 (2009)
Cherny, A., Madan, D.: New measures for performance evaluation. Rev. Financ. Stud. 22, 2571–2606 (2009)
Cherny, A., Madan, D.: Markets as a counterparty: an introduction to conic finance. Int. J. Theor. Appl. Finance 13, 1149–1177 (2010)
Cont, R., Deguest, R., Scandolo, G.: Robustness and sensitivity analysis of risk measurement procedures. Quant. Finance 10, 593–606 (2010)
Chung, K.L.: The strong law of large numbers. In: Neyman, J. (ed.) Proceedings of the Second Berkeley Symposium on Mathematical Statistics and Probability, pp. 341–352. University of California Press, Berkeley (1951)
Cuevas, A.: Qualitative robustness in abstract inference. J. Stat. Plan. Inference 18, 277–289 (1988)
Delbaen, F.: Coherent risk measures. Cattedra Galileiana, Scuola Normale, Superiore, Pisa (2000). http://www.math.ethz.ch/~delbaen/ftp/preprints/RiskMeasuresGeneralSpaces.pdf
Denneberg, D.: Premium calculation: why standard deviation should be replaced by absolute deviation. ASTIN Bull. 20, 181–190 (1990)
Edgar, G.A., Sucheston, L.: Stopping Times and Directed Processes. Cambridge University Press, Cambridge (1992)
El Karoui, N., Ravanelli, C.: Cash subadditive risk measures and interest rate ambiguity. Math. Finance 19, 561–590 (2009)
Farkas, W., Koch-Medina, P., Munari, C.-A.: Beyond cash-additive capital requirements: when changing the numéraire fails. Finance Stoch. 18, 145–174 (2014)
Filipović, D., Svindland, G.: The canonical model space for law-invariant convex risk measures is L 1. Math. Finance 22, 585–589 (2012)
Föllmer, H., Schied, A.: Convex measures of risk and trading constraints. Finance Stoch. 6, 429–447 (2002)
Föllmer, H., Schied, A.: Stochastic Finance. An Introduction in Discrete Time, 3rd edn. de Gruyter, Berlin (2011)
Gilat, D., Helmers, R.: On strong laws for generalized L-statistics with dependent data. Comment. Math. Univ. Carol. 38, 187–192 (1997)
Hampel, F.R.: A general qualitative definition of robustness. Ann. Math. Stat. 42, 1887–1896 (1971)
Huber, P.J., Ronchetti, E.M.: Robust Statistics, 2nd edn. Wiley, New York (2009)
Kallenberg, O.: Foundations of Modern Probability. Springer, New York (1997)
Krätschmer, V., Schied, A., Zähle, H.: Qualitative and infinitesimal robustness of tail-dependent statistical functionals. J. Multivar. Anal. 103, 35–47 (2012)
Kusuoka, S.: On law invariant coherent risk measures. Adv. Math. Econ. 3, 83–95 (2001)
Mizera, I.: Qualitative robustness and weak continuity: the extreme unction. In: Antoch, J., Hušková, M., Sen, P.K. (eds.) Nonparametrics and Robustness in Modern Statistical Inference and Time Series Analysis: A Festschrift in Honor of Professor Jana Jurečková. IMS Collections Festschrift, pp. 169–181. Institute of Mathematical Statistics, Beachwood (2010)
Nelson, D.B.: Stationarity and persistence in the GARCH(1, 1) model. Econom. Theory 6, 318–334 (1990)
Parthasarathy, K.R.: Probability Measures on Metric Spaces. Probability and Mathematical Statistics, vol. 3. Academic Press, New York (1967)
Ruszczynski, A., Shapiro, A.: Optimization of convex risk functions. Math. Oper. Res. 31, 433–451 (2006)
Schied, A.: On the Neyman–Pearson problem for law-invariant risk measures and robust utility functionals. Ann. Appl. Probab. 14, 1398–1423 (2004)
Tsukahara, H.: Estimation of distortion risk measures. J. Financ. Econom. (2013). doi:10.1093/jjfinec/nbt005
van der Vaart, A.W., Wellner, J.A.: Weak Convergence and Empirical Processes. Springer, New York (1996)
van Zwet, W.R.: A strong law for linear functionals of order statistics. Ann. Probab. 8, 986–990 (1980)
Villani, C.: Topics in Optimal Transportation. American Mathematical Society, Providence (2004)
Wang, S.S.: Premium calculation by transforming the layer premium density. ASTIN Bull. 26, 71–92 (1996)
Weber, S.: Distribution-invariant risk measures, information, and dynamic consistency. Math. Finance 16, 419–441 (2006)
Yaari, M.: The dual theory of choice under risk. Econometrica 55, 95–115 (1987)
Zähle, H.: Marcinkiewicz–Zygmund and ordinary strong laws for empirical distribution functions and plug-in estimators. Statistics (2013). doi:10.1080/02331888.2013.800075
Acknowledgements
The authors thank Freddy Delbaen, Paul Embrechts, Marco Frittelli and two anonymous referees for comments, which helped to improve a previous draft of the paper.
Author information
Authors and Affiliations
Corresponding author
Appendix: Auxiliary results on the ψ-weak topology
Appendix: Auxiliary results on the ψ-weak topology
First recall from Corollary A.45 in [21] that the ψ-weak topology on \(\mathcal{M}_{1}^{\psi}(\mathbb{R})\) is separable and metrizable. The following lemma provides some useful characterizations of ψ-weak convergence; see [26, Lemma 3.4] for a proof.
Lemma A.1
The following statements are equivalent:
-
(i)
μ n →μ ψ-weakly.
-
(ii)
∫f dμ n →∫f dμ for every \(f\in C_{\psi}(\mathbb {R})\).
-
(iii)
∫f dμ n →∫f dμ for every continuous f with compact support and for f=ψ.
-
(iv)
μ n →μ weakly and ∫ψ dμ n →∫ψ dμ.
The following lemma gives a transparent characterization of the ψ-weakly compact subsets of \({\mathcal{M}}_{1}^{\psi}\). Recall that a set \({\mathcal{N}}\subset{\mathcal{M}}_{1}^{\psi}\) is called uniformly ψ-integrating if it satisfies (2.12).
Lemma A.2
A set \({\mathcal{N}}\subset{\mathcal{M}}_{1}^{\psi}\) is relatively compact for the ψ-weak topology if and only if there exists a measurable function \(\phi:\mathbb{R}\to[0,\infty)\) such that ϕ(x)/ψ(x)→∞ as |x|→∞ and such that
In this case, \({\mathcal{N}}\) is uniformly ψ-integrating.
Proof
The first statement is an immediate consequence of Corollary A.47 in [21]. For bounded ψ, the second statement is trivial. To prove it for unbounded ψ, we assume without loss of generality that ϕ>0. Fix ε>0, and denote by K the left-hand side of (A.1). Choosing M 1>0 so large that ψ(x)/ϕ(x)≤ε/K when |x|≥M 1, and choosing M 0>0 so large that ψ(x)≥M 0 implies |x|≥M 1, we obtain
for all M≥M 0. That is, (2.12) holds. □
Rights and permissions
About this article
Cite this article
Krätschmer, V., Schied, A. & Zähle, H. Comparative and qualitative robustness for law-invariant risk measures. Finance Stoch 18, 271–295 (2014). https://doi.org/10.1007/s00780-013-0225-4
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00780-013-0225-4
Keywords
- Law-invariant risk measure
- Convex risk measure
- Coherent risk measure
- Orlicz space
- Qualitative robustness
- Comparative robustness
- Index of qualitative robustness
- Hampel’s theorem
- ψ-Weak topology
- Distortion risk measure
- Skorohod representation