Non asymptotic controls on a recursive superquantile approximation
From MaRDI portal
Publication:2233588
DOI10.1214/21-EJS1908zbMATH Open1471.62443arXiv2009.13174OpenAlexW3203193241MaRDI QIDQ2233588FDOQ2233588
Authors: Manon Costa, Sébastien Gadat 
Publication date: 11 October 2021
Published in: Electronic Journal of Statistics (Search for Journal in Brave)
Abstract: In this work, we study a new recursive stochastic algorithm for the joint estimation of quantile and superquantile of an unknown distribution. The novelty of this algorithm is to use the Cesaro averaging of the quantile estimation inside the recursive approximation of the superquantile. We provide some sharp non-asymptotic bounds on the quadratic risk of the superquantile estimator for different step size sequences. We also prove new non-asymptotic -controls on the Robbins Monro algorithm for quantile estimation and its averaged version. Finally, we derive a central limit theorem of our joint procedure using the diffusion approximation point of view hidden behind our stochastic algorithm.
Full work available at URL: https://arxiv.org/abs/2009.13174
Recommendations
- Stochastic approximation algorithms for superquantiles estimation
- Recursive estimation of quantitles using recursive kernel density estimators
- A smoothing stochastic algorithm for quantile estimation
- Central limit theorem and almost sure results for the empirical estimator of superquantiles/CVaR in the stationary case
- Nonparametric recursive quantile estimation
Applications of statistics to actuarial sciences and financial mathematics (62P05) Central limit and other weak theorems (60F05) Stochastic approximation (62L20)
Cites Work
- Coherent measures of risk
- Extreme value theory. An introduction.
- 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
- Stable Paretian models in finance
- Title not available (Why is that?)
- Acceleration of Stochastic Approximation by Averaging
- Title not available (Why is that?)
- Title not available (Why is that?)
- Asymptotic Almost Sure Efficiency of Averaged Stochastic Algorithms
- 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
- Inference in Arch and Garch Models with Heavy-Tailed Errors
- Optimal reinsurance with general risk measures
- Some remarks on the value-at-risk and the conditional value-at-risk
- Analytical value-at-risk with jumps and credit risk
- A modified functional delta method and its application to the estimation of risk functionals
- Additivity properties for value-at-risk under archimedean dependence and heavy-tailedness
- Computing VaR and CVaR using stochastic approximation and adaptive unconstrained importance sampling
- Conditional value-at-risk: optimization approach
- Conditional value-at-risk: semiparametric estimation and inference
- Stochastic approximation with two time scales
- Bregman superquantiles. Estimation methods and applications
- Title not available (Why is that?)
- Expected Utility, Penalty Functions, and Duality in Stochastic Nonlinear Programming
- Convergence rate of linear two-time-scale stochastic approximation.
- Théorèmes de convergence presque sure pour une classe d'algorithmes stochastiques à pas decroissant
- RISK MEASURES FOR NON-INTEGRABLE RANDOM VARIABLES
- Stochastic heavy ball
- Stochastic approximation algorithms for superquantiles estimation
Cited In (2)
This page was built for publication: Non asymptotic controls on a recursive superquantile approximation
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2233588)
