Taming lognormal extremes using deformed exponential and logarithmic functions
Stochastic Environmental Research and Risk Assessment · 2026-11 · 40 (11)
DOI: 10.1007/s00477-026-03362-4 ↗Kurz gefasst
Fachartikel in der Zeitschrift Stochastic Environmental Research and Risk Assessment, online erschienen am 09.11.2026. Verfasst von Dionissios T. Hristopulos und 2 weiteren Autor:innen. Die Arbeit nennt 77 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Technical University of Crete.
Abstract (Originalsprache)
Asymmetric probability distributions are common in natural and engineering datasets. Although the lognormal distribution is a standard modeling choice, it imposes rigid constraints on the behavior of the right tail of both the probability density and the hazard function. We propose a $$\kappa $$ -lognormal distribution constructed using deformed exponential and logarithmic functions. This new model provides a continuous deformation of the lognormal distribution, allowing for lighter right tails governed by a deformation parameter. In certain parameter regimes, the $$\kappa $$ -lognormal distribution can also exhibit bimodality. We derive closed-form expressions for key statistical quantities, including the survival function and hazard rate, as well as moment bounds via hypergeometric functions and series expansions. To facilitate numerical maximum-likelihood estimation (MLE), we provide explicit formulas for the gradient and Hessian of the negative log-likelihood. We also propose a quantile fitting estimation algorithm based on the Wasserstein 2-distance, which leads to a simpler optimization problem than MLE. These results offer new practical tools for modeling skewed data in scientific and engineering applications. Importantly, the weight (heaviness) of the right tail can be controlled by means of the deformation parameter.
Förderung
- Technical University of Crete
Zitieren
APA
Hristopulos, D. T., Baxevani, A., & Kaniadakis, G. (2026). Taming lognormal extremes using deformed exponential and logarithmic functions. Stochastic Environmental Research and Risk Assessment, 40(11). https://doi.org/10.1007/s00477-026-03362-4
Vancouver
Hristopulos DT, Baxevani A, Kaniadakis G. Taming lognormal extremes using deformed exponential and logarithmic functions. Stochastic Environmental Research and Risk Assessment. 2026;40(11). doi:10.1007/s00477-026-03362-4
BibTeX
@article{Hristopulos2026,
author = {Hristopulos, Dionissios T. and Baxevani, Anastassia and Kaniadakis, Giorgio},
title = {Taming lognormal extremes using deformed exponential and logarithmic functions},
journal = {Stochastic Environmental Research and Risk Assessment},
year = {2026},
volume = {40},
number = {11},
doi = {10.1007/s00477-026-03362-4}
}
EUVIMED