A PC-based program is described, the purpose of which is to perform estimation and hypothesis testing for multivariate linear models with options for post-hoc investigation, including discriminant analysis and the determination of the most significant parametric function. The program, which is equally useful as a teaching or a research tool, is based on the formulation of the multivariate general linear model. The user is given complete freedom of expression for models, hypotheses, parametric functions, and data transformations.
Get full access to this article
View all access options for this article.
References
1.
Borland International, Inc. (1987). Turbo Pascal Numerical Methods Toolbox. Scotts Valley, CA: Author.
2.
Lutz, J. G. (1992). A Turbo Pascal unit for approximating the cumulative distribution function of Roy's largest root criterion.Educational and Psychological Measurement, 52, 899-904.
3.
Lutz, J. G. and Cundari, L. A. (1989). Determining the most significant parametric function for a given linear multivariate hypothesis.Joumal of Educational Statistics, 14, 21-28.
4.
Lutz, J. G. and Cundari, L. A. (1990). UGLM: A Pascal program for univariate linear modeling.Educational and Psychological Measurement, 50, 113-115.
