Computation of variance components by the MINQUE method

We present a new method of computing C. R. Rao's MINQUE in variance component models (y = X[beta] + U1[xi]1 + ... + Up[xi]p), which requires only inversion and storage of ni - ni matrices, where ni is the number of columns in Ui. In many cases most of these matrices are of diagonal form. In particular, the derivation of MINQUE equations for univariate nested classification models does not need any inversion of matrices.Variance components computation of MINQUE multivariate models