加权p,q对称熵损失下一类指数分布模型参数的估计

在由信息论中的熵演绎出的一种新损失一加权P,q对称熵损失L(θ,δ)=θ/Pδp+δq/qθq-2(ρ,q>0)下,研究了一类指数分布模型c(x,η)θ-νe-νe-T(x)/θ的参数θ的Bayes估计的一般形式与精确形式,讨论了参数θ的形如cT(X)+d的一类估计的可容许性与不可容许性,并应用积分变换定理证明了参数θ的Bayes估计与可容许估计具有不变性,

Estimator of the Parameter in a Subclass of the Exponential Model under Weighted p, q Symmetric Entropy Loss

For exponential model c（x,n）θ（^-v）e^-T（x）/θ,we dealt with the Bayes estimator and admissible estimator of scale parameterθunder a new loss-weighted p,q symmetric entropy loss L（θ,d） =θ^p/pδ^p+δ^q/qθ^q -2（p,q > 0） which deduced from entropy that related to theory of information.The general and exact form of the Bayes estimator were obtained. The admissibility and inadmissibility of a class of linear estimators of the form cT（X） + d were studied.The invariance of the two estimators was proved using the theorem integral transform.