Home // PESARO 2013, The Third International Conference on Performance, Safety and Robustness in Complex Systems and Applications // View article
Authors:
Daniel Kosiorowski
Keywords: data stream; robust procedure; depth function
Abstract:
Estimation of a conditional distribution is a building block of a variety of statistical procedures used in the modern economics. This estimation is especially difficult in case of an economic data stream, i.e., when data are generated by the multidimensional non-stationary process of unknown form which may contain outliers. In this paper we propose a novel approach for robust monitoring conditional and unconditional distributions in the data streams. Our proposals are based on the idea of adjusted Nadaraya-Watson estimator proposed by Hall et. all (1999) and they appeal to the so called data depth concept. We show very promising statistical properties of our proposals in cases of selected linear and nonlinear data streams models.
Pages: 28 to 31
Copyright: Copyright (c) IARIA, 2013
Publication date: April 21, 2013
Published in: conference
ISSN: 2308-3700
ISBN: 978-1-61208-268-4
Location: Venice, Italy
Dates: from April 21, 2013 to April 26, 2013