2
$\begingroup$

Let the variable $Z$ equal $Z = XY$ where $X$ and $X$ are two i.i.d. continuous random variables which distributions are given by $f_X()$ and $f_Y$. The distribution of $Z$ is given by:

$$f_Z(z) = \int_{-\infty}^\infty f_X(x)f_Y(z/x)\frac{1}{|x|}dx$$

, when $f_X = f_Y = f$ the equation becomes

$$f_Z(z) = \int_{-\infty}^\infty f(x)f(z/x)\frac{1}{|x|}dx$$

If $Z$ is the product of $4$ i.i.d random variables $Z=\prod_{i=1}^4 X_i$, then the distribution of $Z$ is

$$f_Z(z) = \int_{-\infty}^\infty\left(\int_{-\infty}^\infty\left( \int_{-\infty}^\infty f(x_1)f(x_2/x_1)\frac{1}{|x_1|}dx_1\right) f(x_3/x_2)\frac{1}{|x_2|}dx_2\right)f(z/x_3)\frac{1}{|x_3|}dx_3$$

  • Is it correct? If yes, Is there a better way to write this down?

  • What if $Z$ is the product of $n$ i.i.d random variables $Z=\prod_{i=1}^n X_i$? How could we write down the formula?

$\endgroup$
3
$\begingroup$

$$f_Z(z)=\int_{\mathbb R^{n-1}}f(x_1)f(x_2)\cdots f(x_{n-1})f\left(\frac{z}{x_1x_2\cdots x_{n-1}}\right)\frac{\mathrm dx_1\mathrm dx_2\cdots\mathrm dx_{n-1}}{|x_1x_2\cdots x_{n-1}|}$$

$\endgroup$
  • $\begingroup$ Thanks for your answer. Could you please develop a bit on how you found out this formula. Thanks a lot! $\endgroup$ – Remi.b Mar 29 '14 at 13:27
  • $\begingroup$ Yes: change of variables + Jacobian. $\endgroup$ – Did Mar 29 '14 at 13:35

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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