The aim of this paper is to present a method for obtaining a more accu
rate prediction for software reliability growth models (SRGMs). It is
our belief that if we try to use a more general approach implying the
building of a supermodel as a weighted sum of several SRGMs, it may be
possible to obtain more accurate results in prediction. The weight fa
ctors will depend on the values of the prequential likelihood function
s as calculated for each SRGM, the values varying each time a new erro
r is observed. The basic models chosen are the Jelinski-Moranda, Goel-
Okumoto, Duane, Littlewood-Verrall and Keiller-Littlewood models. Fina
lly, we shall compare the SRGMs with the supermodels by using the medi
an estimate and deciding if there are any benefits or constraints in a
pplying this technique.