Abstract

The log-normal distribution is a popular model in biostatistics as in many other fields of statistics. Bayesian inference on the mean and median of the distribution is problematic because, for many popular choices of the prior for variance (on the log-scale) parameter, the posterior distribution has no finite moments, leading to Bayes estimators with infinite expected loss for the most common choices of the loss function. In this paper we propose a generalized inverse Gaussian prior for the variance parameter, that leads to a log-generalized hyperbolic posterior, a distribution for which it is easy to calculate quantiles and moments, provided that they exist. We derive the constraints on the prior parameters that yields finite posterior moments of order r. For the quadratic and relative quadratic loss functions, we investigate the choice of prior parameters leading to Bayes estimators with optimal frequentist mean square error. For the estimation of the lognormal mean we show, using simulation, that the Bayes estimator under quadratic loss compares favorably in terms of frequentist mean square error to known estimators. The theory does not apply only to the mean or median estimation but to all parameters that may be written as the exponential of a linear combination of the distribution's two.