Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Suppose $X$ follows a Poisson distribution with mean $\lambda$ and $Y$ follows an exponential distribution with mean $\displaystyle\frac{1}{\mu}$. How can I find distribution of $Z=XY$?

share|improve this question
2  
I will downvote this question. Please accept answers. You have been told several time in the past to do so. And, looking at the answers you have got, atleast a few of them fully address your questions. –  user21436 Apr 25 '12 at 7:17
1  
I'd just try to calculate it using $\mathbb P_Z(A) = \mathbb P(Z\in A) = \sum_n \mathbb P(XY \in A \mid X= n)\mathbb P(X = n)$ –  martini Apr 25 '12 at 7:18
    
@Sasha Done. Thank You for letting me know. –  user21436 Apr 26 '12 at 0:29

2 Answers 2

up vote 0 down vote accepted

Assuming $X$ and $Y$ are independent, the CDF for $z \ge 0$ is $$ F(z) = P(X Y \le z) = e^{-\lambda} + \sum_{x=1}^\infty e^{-\lambda} \frac{\lambda^x}{x!} (1 - e^{-\mu z/x}) = 1 - \sum_{x=1}^\infty \frac{\lambda^x}{x!} e^{-\lambda-\mu z/x}$$

The sum is very unlikely to be expressible in closed form.

share|improve this answer

Consider $$ F_Z(z) = \mathbb{P}(Z \leqslant z) = \mathbb{P}(X Y \leqslant z) \stackrel{\color\maroon{\text{conditional tower}}}{=} \mathbb{E}\left( \mathbb{P}(X Y \leqslant z |X) \right) = \mathrm{e}^{-\lambda} \mathbf{1}_{z \geqslant 0} + \sum_{x=1}^\infty \frac{\mathrm{e}^{-\lambda}}{x!} F_Y\left( \frac{z}{x} \right) $$

Notice that $F_Z(z)$ has a jump discontinuity at the origin: $$ F_Z(0+) - F_Z(0-) = \mathrm{e}^{-\lambda} $$ meaning that $Z$ is not an absolutely continuous random variable. You can see this rather well of the cumulative function histogram of the simulated product: enter image description here

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.