ABSTRACT
A.1 CONDITIONAL VARIANCES AND COVARIANCES Let F = σ {M, Θ, T} be a σ-algebra, where M is a finite set of models, Θ is a finite set of parameters in the models, and T is a finite set of trends. To simplify the discussion, we may assume that the set of trends is a part of the set of models, that is, T ⊂ M. In this case, we have that F = σ∗ {M, Θ}. This will be discussed in more detail in Section A.2 later.
