English

Asymptotic normality of the deconvolution kernel density estimator under the vanishing error variance

Statistics Theory 2018-04-17 v2 Statistics Theory

Abstract

Let X1,...,XnX_1,...,X_n be i.i.d. observations, where Xi=Yi+σnZiX_i=Y_i+\sigma_n Z_i and the YY's and ZZ's are independent. Assume that the YY's are unobservable and that they have the density ff and also that the ZZ's have a known density k.k. Furthermore, let σn\sigma_n depend on nn and let σn0\sigma_n\to 0 as n.n\to\infty. We consider the deconvolution problem, i.e. the problem of estimation of the density ff based on the sample X1,...,Xn.X_1,...,X_n. A popular estimator of ff in this setting is the deconvolution kernel density estimator. We derive its asymptotic normality under two different assumptions on the relation between the sequence σn\sigma_n and the sequence of bandwidths hn.h_n. We also consider several simulation examples which illustrate different types of asymptotics corresponding to the derived theoretical results and which show that there exist situations where models with σn0\sigma_n\to 0 have to be preferred to the models with fixed σ.\sigma.

Keywords

Cite

@article{arxiv.0807.3540,
  title  = {Asymptotic normality of the deconvolution kernel density estimator under the vanishing error variance},
  author = {Shota Gugushvili and Bert van Es},
  journal= {arXiv preprint arXiv:0807.3540},
  year   = {2018}
}

Comments

22 pages, 8 figures

R2 v1 2026-06-21T11:03:14.488Z