因为
P_Y(y)=F(Y<=y)=F(X^2<=y)={0,y<=0 ; F_X(y^1/2)-F_X(-y^1/2),y>o}
从而
f_Y(y)={0,y<=0 ; f _X(y^1/2)*1/2*y^(-1/2)=1/2*λ*e^(-λ*y^1/2)*y^(-1/2),y>o}
X的分布函数:
F_X(x)={ 1-e^-λx , x>0
{ 0 , x<=0
Y=X^2,X=±√Y
y<=0时,
P{Y<=0}=F_Y(y<=Y) - F_Y(-∞)=F_Y(y<=Y) - 0=0
所以y<=0,F_Y(y)=0,f_Y(y)=F_Y(y)'=0
y>0时,F_Y(y)=F_X(√y)=1-e^(-λ√y)
f_Y(y)=F_Y(y)'=e^(-λ√y)*λ*1/(2√y)=(1/2)λ*y^(-1/2)*e^(-λ√y)
Y的概率密度函数:
f_Y(y)={ (1/2)λ*y^(-1/2)*e^(-λ√y) , y>0
{ 0 , y<=0