Estimation from a censored sample for the exponential family

Abstract

The likelihood equations are derived for the estimation of the parameters of an exponential family from a Type I censored sample and are shown to have an interpretation which suggests a particular iterative method of solution. Some results relevant to the convergence properties of this method are given and the asymptotic variance-covariance matrix is derived. The method is illustrated by an example in which it is compared with an alternative method in the case of estimation for a doubly censored normal distribution.