FAST SELECTION OF BANDWIDTHS FOR NONPARAMETRIC ESTIMATION OF THE PROBABILITY DENSITY OF A TWO-DIMENSIONAL RANDOM VARIABLE WITH DEPENDENT COMPONENTS
A. V. Lapko1,2, V. A. Lapko1,2
1Institute of Computational Modelling, Siberian Branch, Russian Academy of Sciences, Krasnoyarsk, Russia
2Reshetnev Siberian State University of Science and Technology, Krasnoyarsk, Russia
Keywords: nonparametric estimation of the probability density of a two-dimensional random variable, dependent random variables, kernel probability density estimation, fast bandwidth selection
Abstract
A method is proposed for fast selection of bandwidths of kernel functions in the nonparametric estimation of a two-dimensional random variable with dependent components. The method is based on the results of the analysis of the asymptotic properties of the Rosenblatt-Parsen kernel probability density estimation. The properties of a fast algorithm for bandwidth selecting in the considered nonparametric estimation of the probability density are investigated.
|
