Are there any random variables so that E[X] and E[Y] exist but E[XY] doesn't?
Are there any random variables so that E[X] and E[Y] exist but E[XY] doesn't?
1
$\begingroup$
probability
random
-
0You can probably look at some variant of the Cauchy Distribution. – 2011-04-09
-
0what do you mean by $E(XY)$ does not exist? – 2011-04-09
-
3would you be satisfied with probability space $[0,1]$, $X=Y=x^{-1/2}$? – 2011-04-09
-
1@Rasmus That's a different question. – 2011-04-09
-
0@Douglas: Oh, right, thanks you. – 2011-04-10