English

Adaptive and non-adaptive estimation for degenerate diffusion processes

Statistics Theory 2020-02-25 v1 Statistics Theory

Abstract

We discuss parametric estimation of a degenerate diffusion system from time-discrete observations. The first component of the degenerate diffusion system has a parameter θ1\theta_1 in a non-degenerate diffusion coefficient and a parameter θ2\theta_2 in the drift term. The second component has a drift term parameterized by θ3\theta_3 and no diffusion term. Asymptotic normality is proved in three different situations for an adaptive estimator for θ3\theta_3 with some initial estimators for (θ1\theta_1 , θ2\theta_2), an adaptive one-step estimator for (θ1\theta_1 , θ2\theta_2 , θ3\theta_3) with some initial estimators for them, and a joint quasi-maximum likelihood estimator for (θ1\theta_1 , θ2\theta_2 , θ3\theta_3) without any initial estimator. Our estimators incorporate information of the increments of both components. Thanks to this construction, the asymptotic variance of the estimators for θ1\theta_1 is smaller than the standard one based only on the first component. The convergence of the estimators for θ3\theta_3 is much faster than the other parameters. The resulting asymptotic variance is smaller than that of an estimator only using the increments of the second component.

Keywords

Cite

@article{arxiv.2002.10164,
  title  = {Adaptive and non-adaptive estimation for degenerate diffusion processes},
  author = {Arnaud Gloter and Nakahiro Yoshida},
  journal= {arXiv preprint arXiv:2002.10164},
  year   = {2020}
}
R2 v1 2026-06-23T13:51:25.849Z