On Asymptotics Related to Classical Inference in Stochastic Differential Equations with Random Effects

TL;DR

This paper extends the asymptotic analysis of maximum likelihood estimators in stochastic differential equations with random effects, proving strong consistency under weaker conditions and addressing both iid and non-iid cases.

Contribution

It introduces an alternative proof approach for consistency and asymptotic normality, and extends results to non-iid random effects SDE models.

Findings

Proved strong consistency under weaker assumptions.

Established asymptotic normality for non-iid cases.

Extended theoretical results to more general SDE models.

Abstract

Delattre et al. (2013) considered n independent stochastic differential equations (SDEs), where in each case the drift term is associated with a random effect, the distribution of which depends upon unknown parameters. Assuming the independent and identical (iid) situation the authors provide independent proofs of weak consistency and asymptotic normality of the maximum likelihood estimators (MLEs) of the hyper-parameters of their random effects parameters. In this article, as an alternative route to proving consistency and asymptotic normality in the SDE set-up involving random effects, we verify the regularity conditions required by existing relevant theorems. In particular, this approach allowed us to prove strong consistency under weaker assumption. But much more importantly, we further consider the independent, but non-identical set-up associated with the random effects based SDE framework, and prove asymptotic results associated with the MLEs.