the results that have just been presented for estimating the parameter of a bernoulli distribution are also ture for other estimation problems. under fairly general conditions and for a wide class of loss function, the bayes estimators of some parameter