3
$\begingroup$

Let X be a random variable with mean µ and variance $\sigma^2$. Show that

$E(X-\mu)^4\geq \sigma^4$

and use this to show that the kurtosis of $X$ is at least $-2$.

This looks like a form of Chebyshev's equation:

$P(|X-\mu|\geq a) \leq \frac{\sigma^2}{a^2}$

But does not relate the inequality in the probability.

I tried expanding $E(X-\mu)^4$ out, but that seemed like a dead end.

Can someone guide me on how to solve this? If it does deal with Chebyshev, can someone explain in detail how to get the inequality out of the absolute value probability?

$\endgroup$
3
$\begingroup$

Setting $Y=X-\mu$ it must actually be shown that $\mathbb EY^4\geq(\mathbb EY^2)^2$.

This is a direct consequence of: $$\mathbb EY^4-(\mathbb EY^2)^2=\mathsf{Var}Y^2\geq0$$

$\endgroup$
2
$\begingroup$

You don't need Chebyshev. Note that $E(X-\mu)^4=\operatorname{Var}(X-\mu)^2+E^2(X-\mu)^2$. The first term $\ge 0$; the second is $\sigma^4$.

$\endgroup$

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.