Problems of sampling error and accumulated rounding error in canonical variate analysis are discussed. A new technique is presented which appears to be superior to canonical variate analysis when the ratio of variables to sampling units is greater than one to ten. Examples are presented. (Author/JKS)

The accuracy of the varimax and promax methods of rotation of axes in reproducing known factor matrices was examined. It was found that only when all tests are univocal, or nearly so, could one be reasonably confident that an obtained factor matrix faithfully reproduces a contrived matrix. (Author/JKS)

The solution of weakly constrained regression problems typically requires the iterative search, in a given interval, of a point where a certain function has a zero derivative. This note deals with improved bounds for the interval to be searched. (Author)

Illustrates how commonly available structural equation modeling programs can be used to conduct some basic matrix manipulations and generate multivariate normal data with given means and positive definite covariance matrix. Demonstrates the outlined procedure. (SLD)

General algorithms are presented that can be used for optimizing matrix trace functions subject to certain constraints on the parameters. The parameter set that minimizes the majorizing function also decreases the matrix trace function, providing a monotonically convergent algorithm for minimizing the matrix trace function iteratively. (SLD)

Discusses Component Display Theory (CDT) as a method of instructional design, and describes a reorganization of CDT meant to increase its utility in the instructional design process. Matrices in the CDT are explained, and a synthesized matrix based on content and performance elements is provided. (six references) (LRW)

Centering a matrix row-wise and rescaling it column-wise to a unit sum of squares requires an iterative procedure. It is shown that this procedure converges to a stable solution that need not be centered row-wise. The results bear directly on several types of preprocessing methods in Parafac/Candecomp. (Author/TJH)

Presents a matrix of skills necessary for multimedia development. Skills are identified in business, artistic, and technical areas. These skills are then applied to the following reasons for multimedia use: to condense time and material; to effect rapid skills transfer; and to manage feedback and evaluation. (MES)

In matrix comparison, the performance of four vector matching indices (the coefficient of congruence, the Pearson product moment correlation, the "s"-statistic, and kappa) was evaluated. Advantages and disadvantages of each index are discussed, and the performance of each was assessed within the framework of principal components…

Several methodological issues in the multidimensional scaling of coarse dissimilarities were studied, examining whether it was better to scale dissimilarity data directly or to scale a new matrix derived from the original by row comparisons. Findings support an alternative row-comparison measure based on the Jacard coefficient. (SLD)

Presents a fully Bayesian analysis for the Probability Matrix Decomposition (PMD) model using the Gibbs sampler. Identifies the advantages of this approach and illustrates the approach by applying the PMD model to opinions of respondents from different countries concerning the possibility of contracting AIDS in a specific situation. (SLD)

General strategies used to help discover, prove, and generalize identities for Fibonacci numbers are described along with some properties about the determinants of square matrices. A matrix proof for identity (2) that has received immense attention from many branches of mathematics, like linear algebra, dynamical systems, graph theory and others…

In this note the well-known Lagrange identity is extended to matrices. The resulting generalized Lagrange identity is used to give characterizations of symmetry, commutativity of projectors and normality.

In this note 4 x 4 most-perfect pandiagonal magic squares are considered in which rows, columns and the two main, along with the broken, diagonals add up to the same sum. It is shown that the Moore-Penrose inverse of these squares has the same magic property.