Deconvolution for an atomic distribution
Abstract
Let be i.i.d. observations, where and and are independent. Assume that unobservable 's are distributed as a random variable where and are independent, has a Bernoulli distribution with probability of zero equal to and has a distribution function with density Furthermore, let the random variables have the standard normal distribution and let Based on a sample we consider the problem of estimation of the density and the probability We propose a kernel type deconvolution estimator for and derive its asymptotic normality at a fixed point. A consistent estimator for is given as well. Our results demonstrate that our estimator behaves very much like the kernel type deconvolution estimator in the classical deconvolution problem.
Cite
@article{arxiv.0709.3413,
title = {Deconvolution for an atomic distribution},
author = {Bert van Es and Shota Gugushvili and Peter Spreij},
journal= {arXiv preprint arXiv:0709.3413},
year = {2008}
}
Comments
Published in at http://dx.doi.org/10.1214/07-EJS121 the Electronic Journal of Statistics (http://www.i-journals.org/ejs/) by the Institute of Mathematical Statistics (http://www.imstat.org)