0

The question is:

If $E[|X|^k]$ exists, does $E[|X|^m]$ exist for $0 \leq m \leq k$? If so, can you prove it?

Here $X$ is just a generic random variable.

I don't know how to solve this problem. I thought about approaching it by first defining $g(x, i) = |x|^i$ for some $i \in \mathbb{R}^+$.

$$ E[g(x,i)] = \int_{x_{\min}}^{x_{\max}} g(x,i) f_X(x) \\ E[g(x,k)] = \int_{x_{\min}}^{x_{\max}} g(x,k) f_X(x) = c $$

where $c$ is some constant. I don't know if this is the right direction or not? Could someone advise?

1 Answers1

1

Yes. $$ \begin{align*} E[|X|^m] &= \int_{-\infty}^{\infty} |x|^m f_X(x) dx \\ &= \int_{ |x| \leq 1 } |x|^m f_X(x) dx + \int_{ |x| > 1 } |x|^m f_X(x) dx \\ &\leq 1 + \int_{ |x| > 1 } |x|^k f_X(x) dx \leq 1 + E[|X|^k] < \infty \end{align*} $$

Anon
  • 2,460
  • 1
  • 15
  • 23