Advertisement

An EM Algorithm for the Student-t Cluster-Weighted Modeling

  • Salvatore Ingrassia
  • Simona C. Minotti
  • Giuseppe Incarbone
Conference paper
Part of the Studies in Classification, Data Analysis, and Knowledge Organization book series (STUDIES CLASS)

Abstract

Cluster-Weighted Modeling is a flexible statistical framework for modeling local relationships in heterogeneous populations on the basis of weighted combinations of local models. Besides the traditional approach based on Gaussian assumptions, here we consider Cluster Weighted Modeling based on Student-t distributions. In this paper we present an EM algorithm for parameter estimation in Cluster-Weighted models according to the maximum likelihood approach.

Keywords

Local Model Conditional Density Finite Mixture Gaussian Case Gaussian Assumption 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

References

  1. Faria S, Soromenho G (2010) Fitting mixtures of linear regressions. J Stat Comput Simulat 80:201–225MathSciNetzbMATHCrossRefGoogle Scholar
  2. Frühwirth-Schnatter S (2005) Finite mixture and markov switching models. Springer, HeidelbergGoogle Scholar
  3. Gershenfeld N, Schöner B, Metois E (1999) Cluster-weighted modeling for time-series analysis. Nature 397:329–332CrossRefGoogle Scholar
  4. Hurn M, Justel A, Robert CP (2003) Estimating mixtures of regressions. J Comput Graph Stat 12:55–79MathSciNetCrossRefGoogle Scholar
  5. Ingrassia S, Minotti SC, Vittadini G (2010) Local statistical modeling via the cluster-weighted approach with elliptical distributions. ArXiv: 0911.2634v2Google Scholar
  6. Peel D, McLachlan GJ (2000) Robust mixture modelling using the t distribution. Stat Comput 10:339–348CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Salvatore Ingrassia
    • 1
  • Simona C. Minotti
    • 2
  • Giuseppe Incarbone
    • 1
  1. 1.Dipartimento di Impresa, Culture e SocietàUniversità di CataniaCataniaItaly
  2. 2.Dipartimento di StatisticaUniversità di Milano-BicoccaMilanoItaly

Personalised recommendations