对总体回归模型Y=β0+BX+μ,当X=X时 Yo=B+BXo+u 于是 E(X)=E(B+B1X0+1)=B+B1X0+E()=B0+B1X0 而通过样本回归函数=+x,求得拟合值 Yo=Bo+B,xo 的期望为 E(YO)=E(B+B,Xo=E(Bo)+Xoe(Bi)=Bo+BXo Y是个值Y的无偏估计对总体回归模型Y=0+1X+，当X=X0时 Y0 =  0 + 1 X 0 +  于是 0 0 1 0 0 1 0 0 1 0 E(Y ) = E( +  X + ) =  +  X + E() =  +  X 0 0 1 0 0 0 1 0 1 0 ) ˆ ) ( ˆ ) ( ˆ ˆ ) ( ˆ E(Y = E  +  X = E  + X E  =  +  X