Online Statistical Inference for Nonlinear Stochastic Approximation with Markovian Data
Abstract
We study the statistical inference of nonlinear stochastic approximation algorithms utilizing a single trajectory of Markovian data. Our methodology has practical applications in various scenarios, such as Stochastic Gradient Descent (SGD) on autoregressive data and asynchronous Q-Learning. By utilizing the standard stochastic approximation (SA) framework to estimate the target parameter, we establish a functional central limit theorem for its partial-sum process, . To further support this theory, we provide a matching semiparametric efficient lower bound and a non-asymptotic upper bound on its weak convergence, measured in the L\'evy-Prokhorov metric. This functional central limit theorem forms the basis for our inference method. By selecting any continuous scale-invariant functional , the asymptotic pivotal statistic becomes accessible, allowing us to construct an asymptotically valid confidence interval. We analyze the rejection probability of a family of functionals , indexed by , through theoretical and numerical means. The simulation results demonstrate the validity and efficiency of our method.
Cite
@article{arxiv.2302.07690,
title = {Online Statistical Inference for Nonlinear Stochastic Approximation with Markovian Data},
author = {Xiang Li and Jiadong Liang and Zhihua Zhang},
journal= {arXiv preprint arXiv:2302.07690},
year = {2023}
}