Generalized quantiles as risk measures

From MaRDI portal
Publication:2015471





The authors investigate M-quantiles obtained by means of the minimization of an asymmetric convex loss function. The properties of these generalized quantiles as risk measures are discussed. Also, the connection with the zero utility premium principle and with shortfall risk measures introduced by \textit{H. Föllmer} and \textit{A. Schied} [Finance Stoch. 6, No. 4, 429--447 (2002; Zbl 1041.91039)] are pointed out. In particular, the authors show that the only M-quantiles that are coherent risk measures are the expectiles, introduced by \textit{W. K. Newey} and \textit{J. L. Powell} [Econometrica 55, 819--847 (1987; Zbl 0625.62047)] as the minimizers of an asymmetric quadratic loss function. A dual representation of the expectiles as well as their Kusuoka representations are presented. The relationship with CVaR as well as their asymptotic properties for \(\alpha\rightarrow 1\) are analyzed. It is shown that for very heavy tailed distributions expectiles are more conservative than the usual quantiles. Finally, robustness of the expectiles is shown in the sense of Lipschitzianity with respect to the Wasserstein metrics.



Cites work


Cited in
(only showing first 100 items - show all)








This page was built for publication: Generalized quantiles as risk measures

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2015471)