A Monte Carlo Evaluation of Maximum Likelihood Multidimensional Scaling Methods

49 Pages Posted: 10 Dec 1996

See all articles by Tammo H.A. Bijmolt

Tammo H.A. Bijmolt

University of Groningen - Department of Marketing & Marketing Research

Michel Wedel

University of Maryland - Robert H. Smith School of Business, Marketing Department; University of Groningen - Faculty of Economics and Business

Date Written: March 1996

Abstract

We compare three alternative Maximum Likelihood Multidimensional Scaling methods for pairwise dissimilarity ratings, namely MULTISCALE, MAXSCAL, and gurations very well. The recovery of the true dimensionality depends on the test criterion (likelihood ratio test, AIC, or CAIC), as well as on the MLMDS method. The three MLMDS methods test the dissimilarity data equally well. The methods are relatively robust against violations of their distributional assumptions. MULTISCALE outperforms separate Monte Carlo study, it is shown that the MLMDS methods frequently converge to local optima, especially if a random start is used. Rational starts, however, turn out to provide a satisfactory solution for the local optima problem. Implications for researchers intending to apply MLMDS are provided.

JEL Classification: C15, C61, C63

Suggested Citation

Bijmolt, Tammo H.A. and Wedel, Michel, A Monte Carlo Evaluation of Maximum Likelihood Multidimensional Scaling Methods (March 1996). Available at SSRN: https://ssrn.com/abstract=1521 or http://dx.doi.org/10.2139/ssrn.1521

Tammo H.A. Bijmolt (Contact Author)

University of Groningen - Department of Marketing & Marketing Research ( email )

Nettelbosje 2
Groningen, 9747 AE
Netherlands

Michel Wedel

University of Maryland - Robert H. Smith School of Business, Marketing Department ( email )

College Park, MD 20742
United States
301.405.2162 (Phone)
301.405.0146 (Fax)

HOME PAGE: http://www.rhsmith.umd.edu/marketing/faculty/wedel.html

University of Groningen - Faculty of Economics and Business ( email )

Postbus 72
9700 AB Groningen
Netherlands

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
443
Abstract Views
1,978
Rank
120,139
PlumX Metrics