Parallel and bootstrapped stochastic approximation
作者:
Jürgen Dippon,
期刊:
Stochastic Analysis and Applications
(Taylor Available online 1999)
卷期:
Volume 17,
issue 5
页码: 765-798
ISSN:0736-2994
年代: 1999
DOI:10.1080/07362999908809634
出版商: Marcel Dekker, Inc.
关键词: Stochastic Approximation;Parallel Algorithms;Week and Loglog;Invariance Principles in Banach Spaces;Rate of a.s. Convergence;Bootstrapping;Asymptotic Confidence Regions;62L20;62G09;65Y05;60F17;62F25
数据来源: Taylor
摘要:
Parallelization of stochastic approximation procedures can reduce computation and total observation time of a system. Concerning the number of all observations used by the pure sequential and the suggested parallel method a weak invariance principle implies the asymptotic equivalence of both methods. A loglog invariance principle and a rate of a.s. convergence result describe the pathwise properties. Due to the parallel design asymptotic confidence regions can readily be constructed either by computing the bootstrap distribution or the Gaussian limit distribution determined by the empirical covariance
点击下载:
PDF (869KB)
返 回