Multiclus: A New Method for Simultaneously Performing Multidimensional Scaling and Cluster Analysis

Psychometrika, Volume 56, Issue 1, pp 121-136

16 Pages Posted: 1 Jun 2016

See all articles by Wayne S. DeSarbo

Wayne S. DeSarbo

Pennsylvania State University

Daniel J. Howard

Southern Methodist University (SMU) - Marketing Department

Kamel Jedidi

Columbia Business School - Marketing

Date Written: March 1991

Abstract

This paper develops a maximum likelihood based method for simultaneously performing multidimensional scaling and cluster analysis on two-way dominance or profile data. This MULTICLUS procedure utilizes mixtures of multivariate conditional normal distributions to estimate a joint space of stimulus coordinates and K vectors, one for each cluster or group, in a T-dimensional space. The conditional mixture, maximum likelihood method is introduced together with an E-M algorithm for parameter estimation. A Monte Carlo analysis is presented to investigate the performance of the algorithm as a number of data, parameter, and error factors are experimentally manipulated. Finally, a consumer psychology application is discussed involving consumer expertise/experience with microcomputers.

Keywords: multidimensional scaling, cluster analysis, maximum likelihood estimation, consumer psychology

Suggested Citation

DeSarbo, Wayne S. and Howard, Daniel J. and Jedidi, Kamel, Multiclus: A New Method for Simultaneously Performing Multidimensional Scaling and Cluster Analysis (March 1991). Psychometrika, Volume 56, Issue 1, pp 121-136. Available at SSRN: https://ssrn.com/abstract=2787350

Wayne S. DeSarbo (Contact Author)

Pennsylvania State University ( email )

University Park
State College, PA 16802
United States

Daniel J. Howard

Southern Methodist University (SMU) - Marketing Department

United States

Kamel Jedidi

Columbia Business School - Marketing ( email )

New York, NY 10027
United States

Register to save articles to
your library

Register

Paper statistics

Downloads
7
Abstract Views
91
PlumX Metrics