1
$\begingroup$

Let X be a random variable that is uniformly distributed on $[0,1]$. What are the distribution and probability density functions of $Y$ with $Y=\frac{3X}{1-X}$?

I know that the density is the derivative of the distribution function, so if someone could help me find the distribution, I will try to find the density myself. Thanks in advance!

$\endgroup$
  • $\begingroup$ Can you figure out what values $Y$ can take on? $[0,\infty)$? $(-\infty,\infty)$? $[0,1]$? Once you have done that, can you find, for your favorite number $\alpha$, what is $P\{Y \leq \alpha\}$? How about for other, less favorite numbers? Once you have done this, you have found $P\{Y \leq y\}$ from which you can determine the density function. $\endgroup$ – Dilip Sarwate Mar 28 '15 at 16:17
  • $\begingroup$ @DilipSarwate I think $Y $ can take on values from $[0,\infty)$, but I can't really figure out how your next step should be done $\endgroup$ – JimmyP Mar 28 '15 at 16:19
2
$\begingroup$

If $X$ is distributed over $[0,1]$ then $Y$ is distributed over $\mathbb{R}^+$ and for any $r\in\mathbb{R}^+$ we have: $$ \mathbb{P}[Y\leq r]=\mathbb{P}\left[0\leq X \leq \frac{r}{r+3}\right]=\frac{r}{r+3}\tag{1} $$ so, by differentiation, we get that the pdf of $Y$ is given by: $$ f_Y(t) = \frac{3}{(3+t)^2}\cdot\mathbb{1}_{t\geq 0}.\tag{2}$$

$\endgroup$
  • $\begingroup$ Why do we need the indicator function in the pdf? $\endgroup$ – JimmyP Mar 28 '15 at 16:30
  • $\begingroup$ @JimmyP: because $Y$ does not take negative values. $\endgroup$ – Jack D'Aurizio Mar 28 '15 at 16:32
  • $\begingroup$ And how do you know that (1) is valid? $\endgroup$ – JimmyP Mar 28 '15 at 16:41
  • 1
    $\begingroup$ @JimmyP: because if $Y=\frac{3X}{1-X}$, then $X=\frac{Y}{Y+3}$. $\endgroup$ – Jack D'Aurizio Mar 28 '15 at 16:54

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.