曲指数族的欧氏几何方法及其对置信域的应用

Efron and Amari presented a Riemannian geometric framework for qurved exponential families and studied the information loss and the variance of the estimate using this framilies. In this pap propose a relatively rumple geometric frame work in Euclidean space. Based on this new framework, we study eonfidence regiods for curved exponential families which have not been studied by Efron and Amari. The results obtained by Hamilton et al. are extended to curved exponential families.