ABSTRACT
This chapter considers the problem of Bayesian estimation of the parameters of the univariate lognormal distribution. The Bayesian approach to estimation in life testing is motivated by the notion that prior information concerning an unknown life parameter exists, generally in the form of life-test data on prototypes or data on a similar item in an allied product line. The chapter presents Bayesian estimation of the parameters and reliability function with respect to both proper and noninformative prior distributions. It shows that the problem of Bayesian lower bounds on the reliability function. The chapter provides Bayes estimates for several different proper prior distributions on the parameters including conjugate priors and examines noninformative priors. A. W. Drake discussed the appealing aspects of a Bayesian approach to reliability problems, and various lifetime models have been examined in a Bayesian framework.
