A general trimming approach to robust cluster Analysis (original) (raw)
June 2008 A general trimming approach to robust cluster Analysis
Luis A. García-Escudero,Alfonso Gordaliza,Carlos Matrán,Agustin Mayo-Iscar
Ann. Statist. 36(3): 1324-1345 (June 2008). DOI: 10.1214/07-AOS515
Abstract
We introduce a new method for performing clustering with the aim of fitting clusters with different scatters and weights. It is designed by allowing to handle a proportion α of contaminating data to guarantee the robustness of the method. As a characteristic feature, restrictions on the ratio between the maximum and the minimum eigenvalues of the groups scatter matrices are introduced. This makes the problem to be well defined and guarantees the consistency of the sample solutions to the population ones.
The method covers a wide range of clustering approaches depending on the strength of the chosen restrictions. Our proposal includes an algorithm for approximately solving the sample problem.
Citation
Luis A. García-Escudero. Alfonso Gordaliza. Carlos Matrán. Agustin Mayo-Iscar. "A general trimming approach to robust cluster Analysis." Ann. Statist. 36 (3) 1324 - 1345, June 2008. https://doi.org/10.1214/07-AOS515
Information
Published: June 2008
First available in Project Euclid: 26 May 2008
Digital Object Identifier: 10.1214/07-AOS515
Subjects:
Primary: 62H3
Secondary: 62H3
Keywords: asymptotics, cluster analysis, Dykstra’s algorithm, EM-algorithm, fast-MCD algorithm, robustness, trimmed k-means, Trimming
Rights: Copyright © 2008 Institute of Mathematical Statistics
Vol.36 • No. 3 • June 2008
