Effectiveness of Constant Stepsize in Markovian LSA and Statistical Inference

TL;DR

This paper investigates the use of constant stepsize in linear stochastic approximation with Markovian data, demonstrating improved inference procedures, bias reduction techniques, and practical advantages through theoretical and empirical analysis.

Contribution

It introduces a new inference method leveraging constant stepsize LSA, bias reduction via Richardson-Romberg extrapolation, and provides guidance on stepsize selection with theoretical guarantees.

Findings

Constant stepsize improves CI coverage and convergence speed.

Richardson-Romberg extrapolation reduces bias effectively.

Theoretical results guide optimal stepsize selection.

Abstract

In this paper, we study the effectiveness of using a constant stepsize in statistical inference via linear stochastic approximation (LSA) algorithms with Markovian data. After establishing a Central Limit Theorem (CLT), we outline an inference procedure that uses averaged LSA iterates to construct confidence intervals (CIs). Our procedure leverages the fast mixing property of constant-stepsize LSA for better covariance estimation and employs Richardson-Romberg (RR) extrapolation to reduce the bias induced by constant stepsize and Markovian data. We develop theoretical results for guiding stepsize selection in RR extrapolation, and identify several important settings where the bias provably vanishes even without extrapolation. We conduct extensive numerical experiments and compare against classical inference approaches. Our results show that using a constant stepsize enjoys easy hyperparameter tuning, fast convergence, and consistently better CI coverage, especially when data is limited.