【摘要】 设y₁,y₂…,y_n独立同分布,EY₁=β,CovY₁=V,这里βεR^m与V:mxm>0均未知,取损失函数为:L（d,β）=（d-β）′（d-β）,估计类ζ={sum from i=1 to nL_ζY_i+b;L_ζ为m阶实常方阵;i=1,2,…n,bεR^m},本文在损失L下给出了非齐次线估计在ζ中是β的容许估计的充要条件。更多还原

【Abstract】 Suppose Y1,Y2,...Yn ,i.i.d,EY,=β,CovY1 = V,where both BeRm and V: xm>0 are unkown, LetL(d,β) = (d-β)’(d-β)and L is an mxm real matrix ,i=1,2...,n; beRm .In this paper,We give the necessary and sufficient Conditions that a nonhomogeous linearestimator of β is admissible among under L.更多还原