Although the "K"-means algorithm for minimizing the within-cluster sums of squared deviations from cluster centroids is perhaps the most common method for applied cluster analyses, a variety of other criteria are available. The "p"-median model is an especially well-studied clustering problem that requires the selection of "p" objects to serve as…

Data are often a continuous function of a variable such as time observed over some interval. One or more such functions might be observed for each subject. The extension of classical data analytic techniques to such functions is discussed. (Author/JKS)

A variety of problems associated with the interpretation of traditional canonical correlation are discussed. A response surface approach is developed which allows for investigation of changes in the coefficients while maintaining an optimum canonical correlation value. Also, a discrete or constrained canonical correlation method is presented. (JKS)

A revision of the initialization strategy is suggested for OSMOD, a method of nonlinear principal component analysis for multivariate nominal and/or ordinal data. The revision is effective in diminishing the possibility of obtaining locally optimal solutions with categorical data provided by the original version of OSMOD. (SLD)

An extension of Wollenberg's redundancy analysis (an alternative to canonical correlation) is proposed to derive Y-variates corresponding to the optimal X-variates. These variates are maximally correlated with the given X-variates, and depending upon the standardization chosen they also have certain properties of orthogonality. (Author/JKS)

A summary and a unified treatment of fully general computational solutions for two criteria for transforming two or more matrices to maximal agreement are provided. The two criteria--Maxdiff and Maxbet--have applications in the rotation of factor loading or configuration matrices to maximal agreement and the canonical correlation problem. (SLD)

The perturbation theory of the generalized eigenproblem is used to derive influence functions of each squared canonical correlation coefficient and the corresponding canonical vector pair. Three sample versions of these functions are described, and some properties are noted. Two obvious applications, multiple correlation and correspondence…

A multivariate reduced-rank growth curve model is proposed that extends the univariate reduced rank growth curve model to the multivariate case, in which several response variables are measured over multiple time points. The proposed model allows us to investigate the relationships among a number of response variables in a more parsimonious way…

Canonical correlation and redundancy analysis are two approaches to analyzing the interrelationships between two sets of measurements made on the same variables. A component method is presented which uses aspects of both approaches. An empirical example is also presented. (Author/JKS)

A general model is developed for the analysis of multivariate multilevel data structures. Special cases of this model include: repeated measures designs; multiple matrix samples; multilevel latent variable models; multiple time series and variance and covariance component models. (Author)

The frequencies of "m" independent "p-way" contingency tables are analyzed by a model that assumes that the ordinal category data in each of "m" groups are generated from a latent continuous multivariate normal distribution. The model permits analysis of several groups of individuals simultaneously. (TJH)

This paper proposes a generalization of Thurstonian probabilistic choice models for analyzing both multiple preference responses and their relationships. The approach is illustrated by modeling data from two multivariate preference experiments. Preliminary data analyses show that the extension can yield an adequate representation of multivariate…

A general model for two-level multivariate data, with responses possibly missing at random, is described. The model combines regressions on fixed explanatory variables with structured residual covariance matrices. The likelihood function is reduced to a form enabling computational methods for estimating the model to be devised. (Author)

The longitudinal factor analysis" model, which uniquely resolves factors from two occasions of data representing the same persons measured on the same test battery, is shown to be derivable by application of canonical correlation procedures to factor scores. (Author)

Interpretations regarding the effects of exogenous and endogenous variables on endogenous variables in linear structural equation systems depend upon the convergence of a matrix power series. The test for convergence developed by Joreskog and Sorbom is shown to be only sufficient, not necessary and sufficient. (Author/JKS)