Abstracts

A prominence of General Linear Model (GLM) procedure of SAS/STAT (R) software so famous is the feasibility to be adapted to a variety of factorial arrangements, even to a crossed many-factorial one such that not only involving a cell or cells as of missing or plural runs, but also such that a cell or cells omitted or plurified, namely, adapted to the arrangement impaired or extended. Indeterminacy of solutions of the observation equation may be met with special solutions such as of Type I, II, III or IV as typified arbitrarily. Another possible way is such that the design matrix is to be augmented or condensed in the framework by constraints and/or restraints to secure a well-defined determinacy of the solutions as well as the algebraically linear orthogonality as to establish a unified and general scheme of analysis of response and variance