Online estimation of the geometric median in Hilbert spaces: nonasymptotic confidence balls

DOI10.1214/16-AOS1460MaRDI QIDQ2012199zbMATH OpenOpenAlexFDO

Authors Hervé Cardot, Peggy Cénac, Antoine Godichon-Baggioni

Publication date 28 July 2017

Published in The Annals of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1501.06930

zbMATH Keywords

functional data analysisrecursive estimationrobust statisticsspatial medianstochastic gradient algorithmsmartingales in Hilbert spaces

Mathematics Subject Classification ID

Nonparametric estimation (62G05) Nonparametric robustness (62G35) Nonparametric tolerance and confidence regions (62G15) Stochastic approximation (62L20)

Abstract: Estimation procedures based on recursive algorithms are interesting and powerful techniques that are able to deal rapidly with (very) large samples of high dimensional data. The collected data may be contaminated by noise so that robust location indicators, such as the geometric median, may be preferred to the mean. In this context, an estimator of the geometric median based on a fast and efficient averaged non linear stochastic gradient algorithm has been developed by Cardot, C'enac and Zitt (2013). This work aims at studying more precisely the non asymptotic behavior of this algorithm by giving non asymptotic confidence balls. This new result is based on the derivation of improved $L^2$ rates of convergence as well as an exponential inequality for the martingale terms of the recursive non linear Robbins-Monro algorithm.

Recommendations

• Estimating the geometric median in Hilbert spaces with stochastic gradient algorithms: L^p and almost sure rates of convergence
• Recursive estimation of the conditional geometric median in Hilbert spaces
• Efficient and fast estimation of the geometric median in Hilbert spaces with an averaged stochastic gradient algorithm
• Stochastic approximation for multivariate and functional median
• Geometric median and robust estimation in Banach spaces

Cited in

(29)

• Geometric median and robust estimation in Banach spaces
• Evaluating the complexity of some families of functional data
• Convergence in quadratic mean of averaged stochastic gradient algorithms without strong convexity nor bounded gradient
• Non asymptotic controls on a recursive superquantile approximation
• Online bootstrap inference for the geometric median
• Asymptotic properties of high-dimensional spatial median in elliptical distributions with application
• Volatility estimation in a nonlinear heteroscedastic functional regression model with martingale difference errors
• Stochastic approximation algorithms for superquantiles estimation
• Multivariate \(\rho \)-quantiles: a spatial approach
• On the Behavior of Extreme d-dimensional Spatial Quantiles Under Minimal Assumptions
• Halfspace depths for scatter, concentration and shape matrices
• Recursive estimation of the conditional geometric median in Hilbert spaces
• Algorithms of robust stochastic optimization based on mirror descent method
• On the rates of convergence of parallelized averaged stochastic gradient algorithms
• L^p and almost sure rates of convergence of averaged stochastic gradient algorithms: locally strongly convex objective
• Spatial quantiles on the hypersphere
• Conditional quantile sequential estimation for stochastic codes
• The geometric median and applications to robust mean estimation
• Efficient and fast estimation of the geometric median in Hilbert spaces with an averaged stochastic gradient algorithm
• Fast estimation of the median covariation matrix with application to online robust principal components analysis
• Online stochastic Newton methods for estimating the geometric median and applications
• Online estimation of the asymptotic variance for averaged stochastic gradient algorithms
• scientific article; zbMATH DE number 7625199 (Why is no real title available?)
• On weighted multivariate sign functions
• Affine-equivariant inference for multivariate location under \({L_p}\) loss functions
• Estimating the geometric median in Hilbert spaces with stochastic gradient algorithms: \(L^p\) and almost sure rates of convergence
• Stochastic approximation for multivariate and functional median
• An efficient averaged stochastic Gauss-Newton algorithm for estimating parameters of nonlinear regressions models
• Optimal non-asymptotic analysis of the Ruppert-Polyak averaging stochastic algorithm

This page was built for publication: Online estimation of the geometric median in Hilbert spaces: nonasymptotic confidence balls

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2012199)