What is the conditional density function of $Z|Z = X$, where $Z = max\{X, Y\}$, X~exp($\lambda_1$), Y~exp($\lambda_2$) and X, Y are independent?

67 Views Asked by At

Given that $X $~$ $exp$(\lambda_1)$, $Y $~$ $exp$(\lambda_2)$, and $X$,$Y$ are independt.

$Z$=$max\{X,Y\}$

Find the conditional density function of $Z$ given $Z$=$X$.

I get pdf of $Z$ is $\lambda_1 e^{-\lambda_1z} (1 - e^{-\lambda_2z})+\lambda_2 e^{-\lambda_2z} (1 - e^{-\lambda_1z})$

$F_Z(z)=Pr(Z\le z)=Pr(X \le z) \cap Pr(Y \le z) = Pr(X \le z)Pr(Y \le z) = (1 - e^{-\lambda_1z})(1 - e^{-\lambda_2z}), \forall z \ge 0$

Density function of Z is $f_Z(z) = \frac{\partial F_Z(z)}{\partial z} = \lambda_1 e^{-\lambda_1z} (1 - e^{-\lambda_2z})+\lambda_2 e^{-\lambda_2z} (1 - e^{-\lambda_1z})$

But how to write the conditional density function of $Z|Z=X$?

From the definition of Conditional continuous distributions,

$f_Z(z|X=x) = \frac{f_{Z,X}(z,x)}{f_X(x)}$

To fit into this problem, I rewrite the equation above as $f_Z(z|X=z) = \frac{f_{Z,X}(z,z)}{f_X(z)}$. But not sure how to process the next steps.