2
转换订单统计
假设随机变量和是独立的并且是。证明Z_n = n \ log \ frac {\ max(Y _ {(n)},X _ {(n)})} {\ min(Y _ {(n)},X _ {(n)})}的\文本{Exp}(1)分发。X1,...,XnX1,...,XnX_1, ... , X_nY1,...,YnY1,...,YnY_1, ..., Y_nU(0,a)U(0,a)U(0,a)Zn=nlogmax(Y(n),X(n))min(Y(n),X(n))Zn=nlogmax(Y(n),X(n))min(Y(n),X(n))Z_n= n\log\frac{\max(Y_{(n)},X_{(n)})}{\min(Y_{(n)},X_{(n)})}Exp(1)Exp(1)\text{Exp}(1) 我通过设置\ {X_1,...,X_n,Y_1,... Y_n \} = \ {Z_1,...,Z_n \}开始了这个问题,{X1,...,Xn,Y1,...Yn}={Z1,...,Zn}{X1,...,Xn,Y1,...Yn}={Z1,...,Zn}\{X_1,...,X_n,Y_1,...Y_n\} = \{Z_1,...,Z_n\}然后max(Yn,Xn)=Z(2n)max(Yn,Xn)=Z(2n)\max(Y_n,X_n)= Z_{(2n)}分布为(za)2n(za)2n(\frac{z}{a})^{2n}而min(Yn,Xn)=Z(1)min(Yn,Xn)=Z(1)\min(Y_n,X_n)= Z_{(1)}分布为1−(1−za)2n1−(1−za)2n1 - (1 - \frac{z}{a})^{2n} 可以很容易地找到密度,因为fZ1(z)=(2n)(1−za)2n−11afZ1(z)=(2n)(1−za)2n−11af_{Z_{1}}(z) = (2n)(1-\frac{z}{a})^{2n-1}\frac{1}{a}和fZ(2n)(z)=(2n)(za)2n−11afZ(2n)(z)=(2n)(za)2n−11af_{Z_{(2n)}}(z) = (2n)(\frac{z}{a})^{2n-1} \frac{1}{a} 现在,在计算完这些之后,我很难知道下一步要去哪里。我以为它必须进行某种转换,但是我不确定...