Main Article Content

Abstract

Young, Scariano, and Hallum (2005) study two univariate linear regression models, and establish a certain condition for the variances estimates to be equal; however, the condition is in error. They then apply this incorrect condition to some unknown completely randomized design model. The present paper records the correct condition for the equality of the context variances, and points out that this condition cannot be applied to experimental design models.

Keywords

equality of regression parametric estimates equality of the covariances Randomized Blocks design full rank generalized inverse generalized inverse Moore-Penrose inverse

Article Details

How to Cite
Gupta, A. K., & Kabe, D. G. (2011). Covariance Structures of Linear Models. Pakistan Journal of Statistics and Operation Research, 7(2-Sp). https://doi.org/10.18187/pjsor.v7i2-Sp.298