Percentiles of sums of heavy-tailed random variables: beyond the single-loss approximation

DOI10.1007/S11222-013-9376-6zbMATH Open1325.62012arXiv1203.2564OpenAlexW1666677164MaRDI QIDQ892484FDOQ892484

Authors: Lorenzo Hernández, Jorge Tejero, Alberto Suárez, Santiago Carrillo-Menéndez 

Publication date: 19 November 2015

Published in: Statistics and Computing (Search for Journal in Brave)

Abstract: A perturbative approach is used to derive approximations of arbitrary order to estimate high percentiles of sums of positive independent random variables that exhibit heavy tails. Closed-form expressions for the successive approximations are obtained both when the number of terms in the sum is deterministic and when it is random. The zeroth order approximation is the percentile of the maximum term in the sum. Higher orders in the perturbative series involve the right-truncated moments of the individual random variables that appear in the sum. These censored moments are always finite. As a result, and in contrast to previous approximations proposed in the literature, the perturbative series has the same form regardless of whether these random variables have a finite mean or not. The accuracy of the approximations is illustrated for a variety of distributions and a wide range of parameters. The quality of the estimate improves as more terms are included in the perturbative series, specially for higher percentiles and heavier tails.

Full work available at URL: https://arxiv.org/abs/1203.2564

Recommendations

• Maxima of Sums of Heavy-Tailed Random Variables
• Normex, a new method for evaluating the distribution of aggregated heavy tailed risks
• High level quantile approximations of sums of risks
• Series expansions for convolutions of Pareto distributions
• Large deviations for sums of independent heavy-tailed random variables

zbMATH Keywords

heavy tailsvalue at riskaggregate loss distributionsubexponential distributionscensored momentspercentile estimation

Mathematics Subject Classification ID

Censored data models (62N01) Characterization and structure theory of statistical distributions (62E10)

Cites Work

• Title not available (Why is that?)
• Title not available (Why is that?)
• Applied Probability and Queues
• Title not available (Why is that?)
• Heavy-Tail Phenomena
• An Introduction to Heavy-Tailed and Subexponential Distributions
• Title not available (Why is that?)
• Estimates for the probability of ruin with special emphasis on the possibility of large claims
• Title not available (Why is that?)
• Title not available (Why is that?)
• Rare events simulation for heavy-tailed distributions
• Improved algorithms for rare event simulation with heavy tails
• Modeling teletraffic arrivals by a Poisson cluster process
• Title not available (Why is that?)
• Operational Risk
• A review of results on sums of random variables
• Title not available (Why is that?)
• Asymptotic expansions for infinite weighted convolutions of heavy tail distributions and applications
• Second order behaviour of the tail of a subordinated probability distribution
• Higher-order expansions for compound distributions and ruin probabilities with subexponential claims
• Asymptotic expansions of convolutions of regularly varying distributions
• Title not available (Why is that?)
• On subordinated distributions and generalized renewal measures
• Theory prob. appl.
• On the Sums of Independently Distributed Pareto Variates
• The distribution of sums, products and ratios for Lawrance and Lewis's bivariate exponential random variables
• Robust quantification of the exposure to operational risk: bringing economic sense to economic capital
• Asymptotic Expansions for Distributions of Compound Sums of Random Variables with Rapidly Varying Subexponential Distribution
• Title not available (Why is that?)
• Extreme-value analysis of teletraffic data

Cited In (1)

• Normex, a new method for evaluating the distribution of aggregated heavy tailed risks

Uses Software

• QRM

This page was built for publication: Percentiles of sums of heavy-tailed random variables: beyond the single-loss approximation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q892484)