Multidimensional scaling : Some econometric applications

Abstract Multidimensional scaling algorithms seek N points, in M-dimensional space, whose interpoint distances match in some sense the experimental dissimilarities of N variables. The dissimilarity coefficients may be qualitative and of diverse kinds. For M = 2 or 3, complex relationships can be depicted graphically and summarized succinctly. Prevalence of bulky correlation matrices in econometric research suggests multidimensional scaling may prove a helpful tool.