5
$\begingroup$

This is from MITx's Intro to Probability and Statistics course, the problem is on this page.

Suppose $X \sim \textrm{Uniform}(0,1)$ and $Y=X^3$. Find the pdf for $Y$.

Since it's a uniform distribution, I believe the pdf for X is $f(x) = \frac{1}{b-a}$ so in this case just 1. The inverse function of Y is, I think, $g^{-1}(y)=\sqrt[3]{y}$ . The formula I have for finding the pdf for y is $f_y(y)=f_x(g^{-1}(y)) \left|\frac{d}{dy}g^{-1}(y)\right|$. So essentially it seems to be the pdf for y should be the derivative of the inverse function $g^{-1}(y)=\sqrt[3]{y}$.

I've run this many times and I keep getting the answer to be $\frac{1}{3y^{2/3}}$, but the course webpage says the answer is $\frac{1}{3}y^{-2}$. Could anyone explain to me why this is so?

$\endgroup$
  • $\begingroup$ Please add the [self-study] tag & read its wiki. $\endgroup$ – Silverfish Jul 16 '16 at 15:12
5
$\begingroup$

One way to check whether you are right, or the webpage is right is to see if the pdfs integrate to 1. $X$ is defined between 0 and 1, so $Y = X^3$ is also defined between 0 and 1.

$$\int_0^1 \dfrac{1}{3} \dfrac{1}{y^2}dy = \dfrac{1}{3}\left[-\dfrac{1}{y} \right]^1_0 = \text{does not converge}. $$

So clearly the webpage is wrong.

$\endgroup$

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.