A spherical representation of a correlation matrix

It is common practice to perform a principal component analysis (PCA) on a correlation matrix to represent graphically the relations among numerous variables. In such a situation, the variables may be considered as points on the unit hypersphere of an Euclidean space, and PCA provides a sort of best fit of these points within a subspace. Taking into account their particular position, this paper suggests to represent the variables on an optimal three-dimensional unit sphere.

---

Résumé Il est classique d'utiliser une analyse en composantes principales pour représenter graphiquement une matrice de corrélation. Dans une telle situation, les variables peuvent être considérées comme des points sur l'hypersphère unité d'un espace Euclidien, et l'analyse en composantes principales permet d'obtenir une bonne approximation de ces points à l'aide d'un sous-espace Euclidien. Prenant en compte une telle situation géométrique, le présent article suggère de représenter les variables sur une sphère tri-dimensionelle optimale.

     

GENFOLD2: A set of models and algorithms for the general UnFOLDing analysis of preference/dominance data

A general set of multidimensional unfolding models and algorithms is presented to analyze preference or dominance data. This class of models termed GENFOLD2 (GENeral UnFOLDing Analysis-Version 2) allows one to perform internal or external analysis, constrained or unconstrained analysis, conditional or unconditional analysis, metric or nonmetric analysis, while providing the flexibility of specifying and/or testing a variety of different types of unfolding-type preference models mentioned in the literature including Caroll's (1972, 1980) simple, weighted, and general unfolding analysis. An alternating weighted least-squares algorithm is utilized and discussed in terms of preventing degenerate solutions in the estimation of the specified parameters. Finally, two applications of this new method are discussed concerning preference data for ten brands of pain relievers and twelve models of residential communication devices.     

The analysis of free-sorting data: Beyond pairwise cooccurrences

Free-sorting data are obtained when subjects are given a set of objects and are asked to divide them into subsets. Such data are usually reduced by counting for each pair of objects, how many subjects placed both of them into the same subset. The present study examines the utility of a group of additional statistics. the cooccurrences of sets of three objects. Because there are dependencies among the pair and triple cooccurrences, adjusted triple similarity statistics are developed. Multidimensional scaling and cluster analysis — which usually use pair similarities as their input data — can be modified to operate on three-way similarities to create representations of the set of objects. Such methods are applied to a set of empirical sorting data: Rosenberg and Kim's (1975) fifteen kinship terms.The author thanks Phipps Arabie, Lawrence Hubert, Lawrence Jones, Ed Shoben, and Stanley Wasserman for their considerable contributions to this paper.     

Fitting a distance model to homogeneous subsets of variables: Points of view analysis of categorical data

An approach is presented for analyzing a heterogeneous set of categorical variables assumed to form a limited number of homogeneous subsets. The variables generate a particular set of proximities between the objects in the data matrix, and the objective of the analysis is to represent the objects in lowdimensional Euclidean spaces, where the distances approximate these proximities. A least squares loss function is minimized that involves three major components: a) the partitioning of the heterogeneous variables into homogeneous subsets; b) the optimal quantification of the categories of the variables, and c) the representation of the objects through multiple multidimensional scaling tasks performed simultaneously. An important aspect from an algorithmic point of view is in the use of majorization. The use of the procedure is demonstrated by a typical example of possible application, i.e., the analysis of categorical data obtained in a free-sort task. The results of points of view analysis are contrasted with a standard homogeneity analysis, and the stability is studied through a Jackknife analysis.     

The majorization approach to multidimensional scaling for Minkowski distances

The majorization method for multidimensional scaling with Kruskal's STRESS has been limited to Euclidean distances only. Here we extend the majorization algorithm to deal with Minkowski distances with 1≤p≤2 and suggest an algorithm that is partially based on majorization forp outside this range. We give some convergence proofs and extend the zero distance theorem of De Leeuw (1984) to Minkowski distances withp>1.     

A latent class procedure for the structural analysis of two-way compositional data

This paper develops a new procedure for simultaneously performing multidimensional scaling and cluster analysis on two-way compositional data of proportions. The objective of the proposed procedure is to delineate patterns of variability in compositions across subjects by simultaneously clustering subjects into latent classes or groups and estimating a joint space of stimulus coordinates and class-specific vectors in a multidimensional space. We use a conditional mixture, maximum likelihood framework with an E-M algorithm for parameter estimation. The proposed procedure is illustrated using a compositional data set reflecting proportions of viewing time across television networks for an area sample of households.     

Variable selection in clustering

Standard clustering algorithms can completely fail to identify clear cluster structure if that structure is confined to a subset of the variables. A forward selection procedure for identifying the subset is proposed and studied in the context of complete linkage hierarchical clustering. The basic approach can be applied to other clustering methods, too.     

Ordination in the presence of group structure,for general multivariate data

A low-dimensional representation of multivariate data is often sought when the individuals belong to a set ofa-priori groups and the objective is to highlight between-group variation relative to that within groups. If all the data are continuous then this objective can be achieved by means of canonical variate analysis, but no corresponding technique exists when the data are categorical or mixed continuous and categorical. On the other hand, if there is noa-priori grouping of the individuals, then ordination of any form of data can be achieved by use of metric scaling (principal coordinate analysis). In this paper we consider a simple extension of the latter approach to incorporate grouped data, and discuss to what extent this method can be viewed as a generalization of canonical variate analysis. Some illustrative examples are also provided.     

Metric unidimensional scaling and global optimization

For the problem of metric unidimensional scaling, the number of local minima is estimated. For locating the globally optimal solution we develop an approach, called the smoothing technique. Although not guaranteed inevitably to locate the global optimum, the smoothing technique did so in all computational experiments where the global optimum was known.     

Between-group analysis with heterogeneous covariance matrices: The common principal component model

Analysis of between-group differences using canonical variates assumes equality of population covariance matrices. Sometimes these matrices are sufficiently different for the null hypothesis of equality to be rejected, but there exist some common features which should be exploited in any analysis. The common principal component model is often suitable in such circumstances, and this model is shown to be appropriate in a practical example. Two methods for between-group analysis are proposed when this model replaces the equal dispersion matrix assumption. One method is by extension of the two-stage approach to canonical variate analysis using sequential principal component analyses as described by Campbell and Atchley (1981). The second method is by definition of a distance function between populations satisfying the common principal component model, followed by metric scaling of the resulting between-populations distance matrix. The two methods are compared with each other and with ordinary canonical variate analysis on the previously introduced data set.     

先秦云梯形制辨

该文主要从先秦文献中所载之楼车及云梯形制，说明二者实有相同之功能，即用以窥伺敌军。从汉儒服《左传》时引用《兵法》一书，即提供了一项较少为右来学界注意的材料。后人由于二者名称各不相同，乃以二者并不相涉；文中则典籍所载，提出所谓“飞楼”者，当为设于云梯上用来观察敌情之塔楼，以证二者理应极有关系。     

Multidimensional scaling in the city-block metric: A combinatorial approach

We present an approach, independent of the common gradient-based necessary conditions for obtaining a (locally) optimal solution, to multidimensional scaling using the city-block distance function, and implementable in either a metric or nonmetric context. The difficulties encountered in relying on a gradient-based strategy are first reviewed: the general weakness in indicating a good solution that is implied by the satisfaction of the necessary condition of a zero gradient, and the possibility of actual nonconvergence of the associated optimization strategy. To avoid the dependence on gradients for guiding the optimization technique, an alternative iterative procedure is proposed that incorporates (a) combinatorial optimization to construct good object orders along the chosen number of dimensions and (b) nonnegative least-squares to re-estimate the coordinates for the objects based on the object orders. The re-estimated coordinates are used to improve upon the given object orders, which may in turn lead to better coordinates, and so on until convergence of the entire process occurs to a (locally) optimal solution. The approach is illustrated through several data sets on the perception of similarity of rectangles and compared to the results obtained with a gradient-based method.     

A generalization of GIPSCAL for the analysis of nonsymmetric data

Graphical representation of nonsymmetric relationships data has usually proceeded via separate displays for the symmetric and the skew-symmetric parts of a data matrix. DEDICOM avoids splitting the data into symmetric and skewsymmetric parts, but lacks a graphical representation of the results. Chino's GIPSCAL combines features of both models, but may have a poor goodness-of-fit compared to DEDICOM. We simplify and generalize Chino's method in such a way that it fits the data better. We develop an alternating least squares algorithm for the resulting method, called Generalized GIPSCAL, and adjust it to handle GIPSCAL as well. In addition, we show that Generalized GIPSCAL is a constrained variant of DEDICOM and derive necessary and sufficient conditions for equivalence of the two models. Because these conditions are rather mild, we expect that in many practical cases DEDICOM and Generalized GIPSCAL are (nearly) equivalent, and hence that the graphical representation from Generalized GIPSCAL can be used to display the DEDICOM results graphically. Such a representation is given for an illustration. Finally, we show Generalized GIPSCAL to be a generalization of another method for joint representation of the symmetric and skew-symmetric parts of a data matrix.This research has been made possible by a fellowship from the Royal Netherlands Academy of Arts and Sciences to the first author, and by research grant number A6394 to the second author, from the Natural Sciences and Engineering Research Council of Canada. The authors are obliged to Jos ten Berge and Naohito Chino for stimulating comments.     

A special Jackknife for Multidimensional Scaling

In this paper we develop a version of the Jackknife which seems especially suited for Multidimensional Scaling. It deletes one stimulus at a time, and combines the resulting solutions by a least squares matching method. The results can be used for stability analysis, and for purposes of cross validation.     

Computational solutions for the problem of negative saliences and nonsymmetry in INDSCAL

Carroll and Chang have derived the symmetric CANDECOMP model from the INDSCAL model, to fit symmetric matrices of approximate scalar products in the least squares sense. Typically, the CANDECOMP algorithm is used to estimate the parameters. In the present paper it is shown that negative weights may occur with CANDECOMP. This phenomenon can be suppressed by updating the weights by the Nonnegative Least Squares Algorithm. A potential drawback of the resulting procedure is that it may produce two different versions of the stimulus space matrix. To obviate this possibility, a symmetry preserving algorithm is offered, which can be monitored to produce non-negative weights as well. This work was partially supported by the Royal Netherlands Academy of Arts and Sciences.     

Convergence of the majorization method for multidimensional scaling

In this paper we study the convergence properties of an important class of multidimensional scaling algorithms. We unify and extend earlier qualitative results on convergence, which tell us when the algorithms are convergent. In order to prove global convergence results we use the majorization method. We also derive, for the first time, some quantitative convergence theorems, which give information about the speed of convergence. It turns out that in almost all cases convergence is linear, with a convergence rate close to unity. This has the practical consequence that convergence will usually be very slow, and this makes techniques to speed up convergence very important. It is pointed out that step-size techniques will generally not succeed in producing marked improvements in this respect.     

The Young-Householder algorithm and the least squares multidimensional scaling of squared distances

It is shown that replacement of the zero diagonal elements of the symmetric data matrix of approximate squared distances by certain other quantities in the Young-Householder algorithm will yield a least squares fit to squared distances instead of to scalar products. Iterative algorithms for obtaining these replacement diagonal elements are described and relationships with the ELEGANT algorithm (de Leeuw 1975; Takane 1977) are discussed. In large residual situations a penalty function approach, motivated by the ELEGANT algorithm, is adopted. Empirical comparisons of the algorithms are given.An early version of this paper was presented at the Multidimensional Data Analysis Workshop, Pembroke College, Cambridge, July 1985. I want to thank Jan de Leeuw and Yoshio Takane for bringing the ELEGANT algorithm to my attention and for clarifying its rationale and notation. My thanks go also to Stephen du Toit for help with the ALSCAL computations reported in Section 7.     

An evaluation of five algorithms for generating an initial configuration for SINDSCAL

Five different methods for obtaining a rational initial estimate of the stimulus space in the INDSCAL model were compared using the SINDSCAL program for fitting INDSCAL. The effect of the number of stimuli, the number of subjects, the dimensionality, and the amount of error on the quality and efficiency of the final SINDSCAL solution were investigated in a Monte Carlo study. We found that the quality of the final solution was not affected by the choice of the initialization method, suggesting that SINDSCAL finds a global optimum regardless of the initialization method used. The most efficient procedures were the methods proposed by by de Leeuw and Pruzansky (1978) and by Flury and Gautschi (1986) for the simultaneous diagonalization of several positive definite symmetric matrices, and a method based on linearly constraining the stimulus space using the CANDELINC approach developed by Carroll, Pruzansky, and Kruskal (1980).Geert De Soete is supported as Bevoegdverklaard Navorser of the Belgian Nationaal Fonds voor Wetenschappelijk Onderzoek. The authors gratefully acknowledge the helpful comments and suggestions of the reviewers.     

Fitting an extended INDSCAL model to three-way proximity data

The INDSCAL individual differences scaling model is extended by assuming dimensions specific to each stimulus or other object, as well as dimensions common to all stimuli or objects. An alternating maximum likelihood procedure is used to seek maximum likelihood estimates of all parameters of this EXSCAL (Extended INDSCAL) model, including parameters of monotone splines assumed in a quasi-nonmetric approach. The rationale for and numerical details of this approach are described and discussed, and the resulting EXSCAL method is illustrated on some data on perception of musical timbres.