Probability density function of the random variable $Y=X^{3}$

1.2k Views Asked by At

Let $X$ be an absolutely continuous random variable.

I have to find the probability density function of $Y=X^{3}$ in terms of the probability density function of $X$.

To solve it I used the basic formula:

$f_{Y}(y)=f_{X}(g^{-1}(y))\left | \frac{\mathrm{d} g^{-1}(y)}{\mathrm{d} y} \right |$

With $g(x)=x^{3}$.

The problem is that to use this formula I have to verify that the $g$ function is strictly increasing and that the derivative of $g^{-1}$ exists but $\sqrt[3]{y}$ is not derivable at $0$. I want to know if that affects the reasoning. And if I'm wrong, what would be the right way?

1

There are 1 best solutions below

1
On BEST ANSWER

As Henry says in the comments, the issue with the single point $x=0$ does not matter.

In case of doubt, brute-force it:

$$F_Y(y)=P(Y\le y) = P( X^3 \le y)=P(X \le y^{1/3})=F_X(y^{1/3}) \tag 1$$

We've used the property $a^3 \le b \iff a \le b^{1/3}$ (work out the details if not sure).

Finally $$f_Y(y)= \frac{d F_Y(y)}{dy}=f_X(y^{1/3}) \, \frac{1}{3} \, y^{-2/3}$$

This is undefined at $y=0$ (depending on $f_X$, it could have a finite limit), but anyway a single point (more precisely, a set of measure zero) is inconsequential for defining a density probability function.