In the statistical and actuarial literature, Lp-quantiles, p∈[1,+∞), represent an important class of risk measures defined through an asymmetric p-power loss function that generalize the classical (L1-)quantiles. By exploiting inter-order relations between partial moments, we show that for a Student's t distribution with ν∈[1,+∞) degrees of freedom the Lν−j-quantile and the Lj+1-quantile always coincide for any j∈[0,ν−1]. For instance, for a Student's t distribution with 4 degrees of freedom, the L4-quantile and L1-quantile are equal and the same holds for the L3-quantile and L2-quantile; for this distribution, closed form expressions for the Lp-quantile, p=1,2,3,4 are provided. Explicit formulas for the central moments are also established. The usefulness of exact formulas is illustrated on real-world financial data.
Bignozzi, V., Merlo, L., Petrella, L. (2024). Inter-order relations between equivalence for Lp-quantiles of the Student's t distribution. INSURANCE MATHEMATICS & ECONOMICS, 116(May 2024), 44-50 [10.1016/j.insmatheco.2024.02.001].
Inter-order relations between equivalence for Lp-quantiles of the Student's t distribution
Bignozzi, VPrimo
;
2024
Abstract
In the statistical and actuarial literature, Lp-quantiles, p∈[1,+∞), represent an important class of risk measures defined through an asymmetric p-power loss function that generalize the classical (L1-)quantiles. By exploiting inter-order relations between partial moments, we show that for a Student's t distribution with ν∈[1,+∞) degrees of freedom the Lν−j-quantile and the Lj+1-quantile always coincide for any j∈[0,ν−1]. For instance, for a Student's t distribution with 4 degrees of freedom, the L4-quantile and L1-quantile are equal and the same holds for the L3-quantile and L2-quantile; for this distribution, closed form expressions for the Lp-quantile, p=1,2,3,4 are provided. Explicit formulas for the central moments are also established. The usefulness of exact formulas is illustrated on real-world financial data.File | Dimensione | Formato | |
---|---|---|---|
Bignozzi-2024-IME-preprint.pdf
accesso aperto
Tipologia di allegato:
Submitted Version (Pre-print)
Licenza:
Altro
Dimensione
1.8 MB
Formato
Adobe PDF
|
1.8 MB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.