scientific article; zbMATH DE number 3911583
[For the entire collection see Zbl 0566.00022.] Because of their low storage requirements, conjugate gradient algorithms have proved useful in the solution of large and/or sparse linear systems. This paper discusses the application of conjugate gradient algorithms to the models associated with a number of common experimental designs. Under some assumptions about balance, the number of iterations necessary to fit any hierarchical or partial-hierarchical model is less than or equal to the number of terms in the model. The problem of finding an adequate full rank parametrization will also be considered.
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3687601)