Extending Fisher's measure of information

Extending Fisher's measure of information
复制标题

扩展费舍尔的信息测量

DOI:
10.1093/biomet/68.3.695
复制
发表时间:
1981
期刊:
影响因子:
2.7
通讯作者:
P. Ferreira
P. Ferreira
中科院分区:
数学2区
文献类型:
--
作者:
P. Ferreira

文献摘要

被引文献

相似文献

摘要本说明调查可能的扩展费雪的信息量的情况下,有先验知识。参数空间我们假设0 =(a,B),一个真实的开区间,有限或非有限。进一步,我们假设存在适当的先验密度e(0),并且我们用5(0 j x)ocp(x j 0)e(0)表示相应的后验密度函数。关于似然、先验和后验分布的期望分别由E、I0、E0和Eo0I表示。数据的边际密度由q(x)表示。此外,在下文中,我们需要一些正则性条件,基本上是p(x I 0)和δ(0)关于0的可微性以及一些积分和极限的收敛性和可微性。如果δ(x)是0的点估计且当且仅当(x,0)= p(x I 0)e(0),则可以得出
SUMMARY This note investigates possible-extensions of Fisher's measure of information to the case where there is prior knowledge. parameter space. We assume that 0 = (a, b), a real open interval, finite or not. Further, we assume that a proper prior density e(0) exists and we denote by 5(0 j x) oc p(x j 0) e(0) the corresponding posterior density function. The expectations with respect to the likelihood, prior and posterior distributions are denoted by E,I0, Eo and Eo0I, respectively. The marginal density of the data is denoted by q(x). Further, in the following we require some regularity conditions, basically differentiability of p(x I 0) and 6(0) with respect to 0 and convergence and interchangeability of some integrals and limits. If 8(x) is a point estimate of 0 and iff (x, 0) = p(x I 0) e(0), then it follows that