GLM Unique Features - Sigma-Restricted and Overparameterized Models

The STATISTICA General Linear Models module offers the sigma-restricted model and overparameterized model for analyzing (ANOVA) designs with categorical predictor variables. The advantages of each approach are described in the Introductory Overview - Sigma-Restricted vs. Overparameterized Model topic. In short, the sigma-restricted model is much more efficient for factorial ANOVA designs than the overparameterized model; however, the overparameterized model is more flexible when analyzing nested designs or mixed-model ANOVA and ANCOVA designs. Searle (1987) and Searle, Casella, and McCulloch (1992) provide detailed discussions of these models and their advantages, disadvantages, and applications.

Unique features and facilities of GLM:

Copyright © 2021. Cloud Software Group, Inc. All Rights Reserved.