Data Driven Smooth Test for Contaminated Data
In this paper we consider a random variable Y contaminated by an additive noise Z from a known distribution. Our purpose is to test the distribution of the unobserved random variable Y. We propose a data driven statistic based on a nonparametric expansion of the density of Y +Z, which can be applied as well in the continuous case as in the discrete case. The problem is considered at first in the univariate case, and then extended in a multivariate setting with a bootstrap procedure. Finite-sample properties are examined through Monte Carlo and Quasi Monte Carlo simulations in univariate and bivariate cases.
AMS Subject Classification62G10 62F05
Key-wordsBootstrap Contaminated data Data driven test Multivariate sample Orthogonal polynomials Quasi Monte Carlo
Unable to display preview. Download preview PDF.
- Ash, R.B., Gardner, M.F., 1975. Topics in stochastic processes. Probability and Mathematical Statistics, Academic Press, New York.Google Scholar
- L’Ecuyer, P., 2006. Uniform random number generation. In Handbook in OR & MS, Hendersonm, S.G. and Nelson B.L. (Editors), Vol. 13, Chap. 3, Elsevier.Google Scholar