Resumen de On best affine unbiased covariance-preserving prediction of factor scores

Heinz Neudecker

• This paper gives a generalization of results presented by ten Berge, Krijnen,Wansbeek & Shapiro. They examined procedures and results as proposed by Anderson & Rubin, McDonald, Green and Krijnen, Wansbeek & ten Berge.We shall consider the same matter, under weaker rank assumptions. We allow some moments, namely the variance O of the observable scores vector and that of the unique factors, ?, to be singular. We require T' ? T > 0, where T ? T' is a Schur decomposition of O. As usual the variance of the common factors, F, and the loadings matrix A will have full column rank.