Template-Type: ReDIF-Paper 1.0
Author-Name: van de Velden, M.
Author-Name-Last: van de Velden
Author-Name-First: Michel
Author-Name: Iodice D' Enza, A.
Author-Name-Last: Iodice D' Enza
Author-Name: Palumbo, F.
Author-Name-Last: Palumbo
Title: Cluster Correspondence Analysis
Abstract: __Abstract__
 
 A new method is proposed that combines dimension reduction and cluster 
 analysis for categorical data. A least-squares objective function is 
 formulated that approximates the cluster by variables cross-tabulation. 
 Individual observations are assigned to clusters in such a way that the 
 distributions over the categorical variables for the different clusters 
 are optimally separated. In a unified framework, a brief review of 
 alternative methods is provided and performance of the methods is 
 appraised by means of a simulation study. The results of the joint 
 dimension reduction and clustering methods are compared with cluster 
 analysis based on the full dimensional data. Our results show that the 
 joint dimension reduction and clustering methods outperform, both with 
 respect to the retrieval of the true underlying cluster structure and 
 with respect to internal cluster validity measures, full dimensional 
 clustering. The differences increase when more variables are involved 
 and in the presence of noise variables.
Length: 30
Creation-Date: 2014-10-01
File-URL: https://repub.eur.nl/pub/77010/EI2014-24.pdf
File-Format: application/pdf
Series: RePEc:ems:eureir
Number: EI 2014-24
Keywords: Correspondence analysis, cluster analysis, dimension, reduction, categorical variables
Handle: RePEc:ems:eureir:77010