English

Moderate Deviation Principle for a Stochastic Approximation Process

Probability 2026-05-11 v1

Abstract

In this paper, we investigate a stochastic approximation procedure (Xn)n0\left(X_n\right)_{n\ge 0} taking values in RR. The process is adapted to a filtration (Fn)n0(F_n)_{n\ge 0} and satisfies the recursion Xn+1=Xn+bn+1[g(Xn)+Un+1]X_{n+1}=X_n+\frac{b}{n+1}\big[g(X_n)+U_{n+1}\big], where b>0b>0, g:RRg:R \to R is a function and (Un)n1\left(U_n\right)_{n\ge 1} is a sequence of bounded martingale differences adapted to the filtration (Fn)n1(F_n)_{n\ge 1}. We establish the moderate deviation principle for the stochastic process (Xn)n0(X_n)_{n\ge 0}. As auxiliary results, we also obtain the exponential inequality for (Xn)n0(X_n)_{n\ge 0} and the moderate deviation principle for weighted sums of bounded martingale differences.

Keywords

Cite

@article{arxiv.2605.07369,
  title  = {Moderate Deviation Principle for a Stochastic Approximation Process},
  author = {Jianan Shi and Qing Yin and Yu Miao},
  journal= {arXiv preprint arXiv:2605.07369},
  year   = {2026}
}

Comments

22 pages, 0 figures

R2 v1 2026-07-01T12:57:06.693Z