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.

Given two independent random variables X and Y, how can I find the PDF of random variable $Z=XY$?

*If their joint distribution is required, assume that we also have it.

share|improve this question
2  
Re the asterisk: independence implies that the joint distribution is fully determined by the marginals. –  Did Sep 17 '13 at 6:26
    
Please do not modify your posts making answers or comments appear off-topic. –  Did Sep 23 '13 at 17:53
    
@Did: you are right. Thanks for your editing. –  May Sep 23 '13 at 17:54

2 Answers 2

up vote 3 down vote accepted

If $X$ and $Y$ are two random variables, you may find the product distribution as follows: $$ f_Z(z)=\int_{-\infty}^{\infty} \frac{1}{|t|} f_X\left(t\right)f_Y\left(\frac{z}{t}\right)dt $$ To see this, suppose that the distribution of $X$ is continuous at 0: $$ P(Z\leq z)=P(XY\leq z)= P(Y\leq \frac{z}{X}\big|X>0)P(X>0)+P(Y\geq \frac{z}{X}\big|X<0)P(X<0)= $$ $$ =\int_{0}^{\infty} P(Y\leq \frac{z}{t}\big)f_X\left(t\right)dt+ \int_{-\infty}^{0} P(Y\geq \frac{z}{t}\big) f_X\left(t\right)dt $$ We can find the derivation of both sides w.r.t. $z$ and we get:

$$ f_Z(z)=\int_{0}^{\infty} \frac{1}{t} f_Y(\frac{z}{t}\big)f_X\left(t\right)dt+ \int_{-\infty}^{0} \frac{-1}{t} f_Y(\frac{z}{t}\big) f_X\left(t\right)dt $$ $$ =\int_{-\infty}^{\infty} \frac{1}{|t|} f_X\left(t\right)f_Y\left(\frac{z}{t}\right)dt $$ The moral: it is not always possible to use this formula.

share|improve this answer
    
thanks, just a question why do we need $1/|t|$? –  May Sep 16 '13 at 23:25
1  
I wrote a more complete answer :-) –  Arash Sep 16 '13 at 23:39
    
Thanks a lot now I see why. I think the last line of formula should be edited. –  May Sep 17 '13 at 0:06
    
just a question why is it not always possible to use it? –  May Sep 17 '13 at 0:07
    
Sorry but I do not understand "it is not always possible to use this formula". The formula is always right. –  Did Sep 17 '13 at 8:58

\begin{align} {\cal P}\left(Z\right) &= {{\rm d} \over {\rm d}Z}\int_{-\infty}^{Z}{\cal P}\left(Z'\right)\,{\rm d}Z' = {{\rm d} \over {\rm d}Z}\int_{-\infty}^{\infty}\int_{-\infty}^{\infty} {\rm P_{X}}\left(X\right)\,{\rm P_{Y}}\left(Y\right)\, \Theta\left(Z - XY\right)\,{\rm d}X\,{\rm d}Y\, \\[3mm]&= \int_{-\infty}^{\infty}\int_{-\infty}^{\infty} {\rm P_{X}}\left(X\right)\,{\rm P_{Y}}\left(Y\right)\, \delta\left(Z - XY\right)\,{\rm d}X\,{\rm d}Y\, = \int_{-\infty}^{\infty}\int_{-\infty}^{\infty} {\rm P_{X}}\left(X\right)\,{\rm P_{Y}}\left(Y\right)\, {\delta\left(Y - Z/X\right) \over \left\vert X\right\vert}\,{\rm d}X\,{\rm d}Y\, \end{align}

$$ \begin{array}{|c|}\hline\\ \color{#ff0000}{\large\quad% {\cal P}\left(Z\right) = \int_{-\infty}^{\infty} {\rm P_{X}}\left(X\right)\,{\rm P_{Y}}\left(Z \over X\right)\, \,{{\rm d}X \over \left\vert X\right\vert} \quad} \\ \\ \hline \end{array} $$

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.