论文部分内容阅读
A general method for assessing local influence of minor perturbations of prior in Bayesian analysis is developed in this paper. Using some elementary ideas from differential geometry, we provide a unified approach for handling a variety of problems of loc