Mixture Models based on Multiresolution Analysis Theory

by F. Palacios-González and R. García-Fernandez.

Abstract: The aim of this paper is to define a new family of probability density functions (MR pdf) based on the multiresolution analysis theory. Each function of this family can be seen as a particular type of density mixture. The MR pdf has advantages of estimation over the conventional mixtures and it is suitable to model a large variety of square integrable probability density functions. The flexibility and the properties of local analysis of the MR pdf facilitate the modelling of the splitting and merging of sub-populations into a given population. In order to do this two algorithms are provided.

Key Words: cubic box spline; income distribution; location of sub-populations; mixture model; multiresolution analysis

F. Palacios-González, fpalacio@ugr.es
Rosa García-Fernández, rosamgf@ugr.es

Editor: Ghorai, Jugal K, jugal@csd.uwm.edu

READING THE ARTICLE: You can read the article in portable document (.pdf) format (343815 bytes.)

NOTE: The content of this article is the intellectual property of the authors, who retains all rights to future publication.

This page has been accessed 1919 times since FEBRUARY 8, 2012.

Return to the InterStat Home Page.