Inference for L orthogonal models

Abstract

A mixed model Yo = ∑m i=1 Xiβi + ∑i=m+1 w Xiβ̃ + e is orthogonal when the matrices Mi = XiXi T, i = 1,...,w, commute. The vectors βc1 1,...,βcm m are fixed vectors and the βcm+1 m+1,...,βcw w and en are random. For these models we have very interesting results namely we have UMVUE for the relevant parameters when normality is assumed. We now intend to generalize that class of models taking Y = L(∑i=1 mXiβi + ∑i=m+1 w Xiβ̃ + e with L a matrix whose column vectors are linearly independent.