1
$\begingroup$

Prove that $E[x^n] \geq (Ex)^n$ for $n = 2k$

I only have the formula of E(x) but I don't know how to prove it.

$\endgroup$
4
  • 2
    $\begingroup$ Welcome to Cross Validated! Can you use Jensen’s inequality? $\endgroup$
    – Dave
    Feb 7, 2022 at 10:57
  • $\begingroup$ Maybe you can try mathematical induction? $\endgroup$
    – toni057
    Feb 7, 2022 at 11:09
  • 1
    $\begingroup$ Please add the self-study tag & read its wiki. Then tell us what you understand thus far, what you've tried & where you're stuck. We'll provide hints to help you get unstuck. Please make these changes as just posting your homework & hoping someone will do it for you is grounds for closing. $\endgroup$ Feb 7, 2022 at 13:39
  • $\begingroup$ 1. Prove that all power functions of the type $f(x) = x^n ,n \epsilon N$ are convex functions. 2. Use Jensen’s inequality as pointed out by Dave $\endgroup$
    – Romain
    Feb 7, 2022 at 14:18

1 Answer 1

1
$\begingroup$

First recall that $\text{Var}(X) = \mathbb{E}[X^2] - \mathbb{E}[X]^2$ and that $\text{Var}(X) \geq 0$.

So $\mathbb{E}[X^{2k}] = \mathbb{E}[(X^k)^2] = \text{Var}(X^k) + \mathbb{E}[X^k]^2.$

Since $\text{Var}(X^k) \geq 0$, $\, \, \, \, \mathbb{E}[X^{2k}] \geq \mathbb{E}[X^k]^2$

Edit: Just realised this doesn't answer your question. This only works when $n$ is a power of 2 (by repeating the argument multiple times). It doesn't work for all even $n$

$\endgroup$
1
  • 1
    $\begingroup$ Why does this not work for all even (positive) $n$? Perhaps you are contemplating an inductive argument, but there are other approaches. For instance, $n$ needn't be integral at all: you should be analyzing $|X|$ rather than $X.$ The only reason in the question to write $2k$ was to make it so that $X^{2k}=|X|^{2k}.$ $\endgroup$
    – whuber
    Feb 7, 2022 at 16:16

Not the answer you're looking for? Browse other questions tagged or ask your own question.