Central limit theorem and Cram\'{e}r-type moderate deviations for Milstein scheme

Abstract

In this paper, we investigate the Milstein numerical scheme with step size $\eta$ for a stochastic differential equation driven by multiplicative Brownian motion. Under some appropriate coefficient conditions, the continuous-time system and its discrete Milstein scheme approximation each possess unique invariant measures, which we denote by $\pi$ and $\pi_\eta$ respectively. We first establish a central limit theorem for the empirical measure $\Pi_{\eta}$, a statistical consistent estimator of $\pi_{\eta}$. Subsequently, we derive both normalized and self-normalized Cram\'{e}r-type moderate deviations.