Asymptotic confidence regions of stochastic approximation procedures in Hilbert spaces

DOI10.1007/BF01046993zbMath0721.62080MaRDI QIDQ2641050

Jürgen Dippon

Publication date: 1991

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

zbMATH Keywords

separable Hilbert space; stopping rules; Gaussian limit distribution; Robbins-Monro method; asymptotic confidence balls; centered Gaussian random variables; Donsker type invariance principle; estimator of the unknown radius of a ball; random covariance operators; sequence of empirical covariance operators

Mathematics Subject Classification ID

62L20: Stochastic approximation

62L15: Optimal stopping in statistics

Related Items

Parallel and bootstrapped stochastic approximation, Recursive estimation: asymptotic confidence regions by empirical quantiles

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• An invariance principle for a finite dimensional stochastic approximation method in a Hilbert space
• Minimizing noisy functionals in Hilbert space: An extension of the Kiefer-Wolfowitz procedure
• Berry-Esseen estimates in Hilbert space and an application to the law of the iterated logarithm
• Martingales and the Robbins-Monro procedure in \(D[0,1\)]
• Bayesian inferences related to shifting sequences and two-phase regression
• Functional and random central limit theorems for the Robbins-Munro process
• On the Asymptotic Theory of Fixed-Size Sequential Confidence Bounds for Linear Regression Parameters
• On the Asymptotic Theory of Fixed-Width Sequential Confidence Intervals for the Mean
• Fixed Size Confidence Ellipsoids for Linear Regression Parameters
• A limit theorem for the Robbins-Monro approximation