In the linear model Y = Xβ+ε with X having a full column rank, all β parameters can be estimated and the estimates are unique. However, in cases where X does not have a full column rank, not all β parameters can be estimated. In this paper, the problem to be discussed is how to determine parameters or parameter functions that are estimable and testable. Applications to the case of unbalanced data will be presented.
Copyrights © 2025