English

Adaptive density estimation for general ARCH models

Statistics Theory 2016-08-16 v1 Statistics Theory

Abstract

We consider a model Y_t=σ_tη_tY\_t=\sigma\_t\eta\_t in which (σ_t)(\sigma\_t) is not independent of the noise process (η_t)(\eta\_t), but σ_t\sigma\_t is independent of η_t\eta\_t for each tt. We assume that (σ_t)(\sigma\_t) is stationary and we propose an adaptive estimator of the density of ln(σ2_t)\ln(\sigma^2\_t) based on the observations Y_tY\_t. Under various dependence structures, the rates of this nonparametric estimator coincide with the minimax rates obtained in the i.i.d. case when (σ_t)(\sigma\_t) and (η_t)(\eta\_t) are independent, in all cases where these minimax rates are known. The results apply to various linear and non linear ARCH processes.

Keywords

Cite

@article{arxiv.math/0609745,
  title  = {Adaptive density estimation for general ARCH models},
  author = {Fabienne Comte and Jérôme Dedecker and Marie-Luce Taupin},
  journal= {arXiv preprint arXiv:math/0609745},
  year   = {2016}
}