ABSTRACT
In this chapter we review some basic results from linear model theory, including least squares estimation, the Gauss-Markov Theorem, and tests of linear hypotheses, with applications to the analysis of fixed-effect one-way and additive two-way layout models. We show how the analysis of an additive two-way layout model is simplified when certain vector spaces associated with the two factors are orthogonal. This geometric condition is shown to be equivalent to that the two factors satisfy the condition of proportional frequencies.
