Stochastic approximation algorithms for superquantiles estimation

DOI10.1214/21-EJP648zbMATH Open1469.62323arXiv2007.14659MaRDI QIDQ2042800FDOQ2042800

Authors: Bernard Bercu, Manon Costa, Sébastien Gadat 

Publication date: 21 July 2021

Published in: Electronic Journal of Probability (Search for Journal in Brave)

Abstract: This paper is devoted to two different two-time-scale stochastic approximation algorithms for superquantile estimation. We shall investigate the asymptotic behavior of a Robbins-Monro estimator and its convexified version. Our main contribution is to establish the almost sure convergence, the quadratic strong law and the law of iterated logarithm for our estimates via a martingale approach. A joint asymptotic normality is also provided. Our theoretical analysis is illustrated by numerical experiments on real datasets.

Full work available at URL: https://arxiv.org/abs/2007.14659

Recommendations

• Non asymptotic controls on a recursive superquantile approximation
• Central limit theorem and almost sure results for the empirical estimator of superquantiles/CVaR in the stationary case
• Superquantile/CVaR risk measures: second-order theory
• Bregman superquantiles. Estimation methods and applications
• A smoothing stochastic algorithm for quantile estimation

zbMATH Keywords

stochastic approximationlimit theoremsconditional value-at-riskquantile and superquantile

Mathematics Subject Classification ID

Nonparametric regression and quantile regression (62G08) Applications of statistics to actuarial sciences and financial mathematics (62P05) Central limit and other weak theorems (60F05) Stochastic approximation (62L20)

Cites Work

• A Note on Quantiles in Large Samples
• Online estimation of the geometric median in Hilbert spaces: nonasymptotic confidence balls
• Estimating the geometric median in Hilbert spaces with stochastic gradient algorithms: \(L^p\) and almost sure rates of convergence
• Acceleration of Stochastic Approximation by Averaging
• Title not available (Why is that?)
• Title not available (Why is that?)
• A Stochastic Approximation Method
• Convergence rate and averaging of nonlinear two-time-scale stochastic approximation algo\-rithms
• Efficient and fast estimation of the geometric median in Hilbert spaces with an averaged stochastic gradient algorithm
• Adaptive design and stochastic approximation
• A New Proof of the Bahadur Representation of Quantiles and an Application
• Title not available (Why is that?)
• On Asymptotic Normality in Stochastic Approximation
• Title not available (Why is that?)
• Computing VaR and CVaR using stochastic approximation and adaptive unconstrained importance sampling
• On a Stochastic Approximation Method
• Stochastic approximation with two time scales
• On the almost sure asymptotic behaviour of stochastic algorithm
• On the convergence of moments in the almost sure central limit theorem for martingales with statistical applications
• Asymptotic Distribution of Stochastic Approximation Procedures
• Online estimation of the asymptotic variance for averaged stochastic gradient algorithms
• Expected Utility, Penalty Functions, and Duality in Stochastic Nonlinear Programming
• On the Law of the Iterated Logarithm in Stochastic Approximation Processes
• Convergence rate of linear two-time-scale stochastic approximation.
• Almost sure approximation of the Robbins-Monro process by sums of independent random variables
• Stochastic heavy ball

Cited In (6)

• A smoothing stochastic algorithm for quantile estimation
• Non asymptotic controls on a recursive superquantile approximation
• Superquantile regression with applications to buffered reliability, uncertainty quantification, and conditional value-at-risk
• Bregman superquantiles. Estimation methods and applications
• Inducing Super-Approximation
• Central limit theorem and almost sure results for the empirical estimator of superquantiles/CVaR in the stationary case

This page was built for publication: Stochastic approximation algorithms for superquantiles estimation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2042800)