R. Thompson et al., CALCULATION OF PREDICTION ERROR VARIANCES USING SPARSE-MATRIX METHODS, Journal of animal breeding and genetics, 111(2), 1994, pp. 102-109
The use of exact and approximate algorithms to calculate prediction er
ror variances using sparse matrix methods are demonstrated for an indi
vidual animal effect including maternal effects. One exact algorithm i
s substantially faster than two others. An approximation of the best e
xact method gave an acceptable level of reliabilities and reduced the
computation by a factor of approximately fifty compared with the exact
computation and is routine in national beef evaluation in Britain.