English

Estimating discontinuous periodic signals in a non-time homogeneous diffusion process

Statistics Theory 2010-03-18 v2 Statistics Theory

Abstract

We consider a diffusion (ξt)t0(\xi_t)_{t\ge 0} with some TT-periodic time dependent input term contained in the drift: under an unknown parameter \vthΘ\vth\in\Theta, some discontinuity - an additional periodic signal - occurs at times kT+\vthkT{+}\vth, k\bbnk\in\bbn. Assuming positive Harris recurrence of (ξkT)k\bbn0(\xi_{kT})_{k\in\bbn_0} and exploiting the periodicity structure, we prove limit theorems for certain martingales and functionals of the process (ξt)t0(\xi_t)_{t\ge 0}. They allow to consider the statistical model parametrized by \vthΘ\vth\in\Theta locally in small neighbourhoods of some fixed \vth\vth, with radius 1/n1/n as \nto\nto. We prove convergence of local models to a limit experiment studied by Ibragimov and Khasminskii [IH 81] and discuss the behaviour of estimators under contiguous alternatives.

Keywords

Cite

@article{arxiv.0903.5061,
  title  = {Estimating discontinuous periodic signals in a non-time homogeneous diffusion process},
  author = {Reinhard Hoepfner and Yury Kutoyants},
  journal= {arXiv preprint arXiv:0903.5061},
  year   = {2010}
}

Comments

42 pages

R2 v1 2026-06-21T12:45:48.270Z