Abstract
Relation between metric multidimensional scaling and principal component analysis. The fundamental relations between metric multidimensional scaling (MDS-M) and principal component analysis (PCA) are explained. Both models are described in terms of type of data matrix analyzed, fundamental assumptions made, computational procedures, solutions and results. It is emphasized that the principal component analysis (ACP) is considered a R-technique, operating on covariance matrices, whereas the metric multidimensional scaling (MDSM) is a Q-technique which deal on matrices consisting of association taken to be a fundamental identity between the two models is shown to be merely the employment of the same theorems to arrive the solution.
