Is there a random variable $X$ such that $E[ |X| ] = E [ X^2 ]$?

155 Views Asked by At

In probability, is it possible for some random variable $X$, that $E[|X|]=E[X^2]$? Can you explain why as well?

I believe the answer has something to do with symmetric distributions but I'm not 100% sure.

1

There are 1 best solutions below

0
On BEST ANSWER

Cauchy-Schwarz inequality gives $$ \mathbb E\lvert X\rvert\leqslant \left(\mathbb E\left[X^2\right]\right)^{1/2} $$ so the condition in the question gives $\mathbb E\left[X^2\right]\leqslant 1$ and not so much information.

Random variable taking the values $0$ and $1$ satisfy $\mathbb E\lvert X\rvert=\mathbb E\left[X^2\right]$ since $\lvert X\rvert=X^2$, hence indicator functions are examples.