If there exists corelation between X and Y, then finding the variance of Z becomes slightly more complicated due to the requirement of knowing the covariance of X and Y.
Syntax: [] = covariance, <> = variance, "" = Mean, () = Normal Paranthese
<Z> = [XY] . ( [XY] + 2."X"."Y" ) + <X>.<Y> + <X>("Y"^2) + <Y>("X"^2)
Hope it helps.
{Btw, it can be simplified into just 3 terms: <Z> = "XY"^2 + ("XX"."YY") - (2 . ("X"^2) . ("Y"^2)), but as I said you need to take covariance into account. }