4
$\begingroup$

Now I know that the variance of the product of two independent variables $Y$ and $Z$ is:$\DeclareMathOperator{\Var}{Var}$

$\Var(YZ) = \Var(Y)\Var(Z) + \Var(Y)E(Z)^2+\Var(Z)E(Y)^2$

However I would like to know what the variance of the product of $Y\mid X$ and $Z\mid X$ is (assuming they are independent as well)? Is it the same as above? Which in turns give me:

$\Var(YZ\mid X) = \Var(Y\mid X) \Var(Z\mid X) + \Var(Y\mid X) E(E[Z\mid X])^2 + \Var(Z\mid X) E(E[Y\mid X])^2$

where I know that applying total expectation law by $E(E[Z\mid X]) = E(Z)$ such that I get:

$\Var(YZ\mid X) = \Var(Y\mid X) \Var(Z\mid X) + \Var(Y\mid X)E(Z)^2 + \Var(Z\mid X)E(Y)^2$

or do I have to apply the total variance law somehow?

$\Var(Y) = E[\Var(Y\mid X)] + \Var[E(Y\mid X)]$

Or is it some other method I have use to get the desired equation? Any help or input is appreciated!

$\endgroup$
5
$\begingroup$

We can follow a similar derivation, assuming conditional independence of $Y$ and $Z$ given $X$: $$\begin{align}\operatorname{var}(YZ|X)&=E[Y^2|X]E[Z^2|X]-E[Y|X]^2E[Z|X]^2\\&=(\operatorname{var}(Y|X)+E[Y|X]^2)(\operatorname{var(Z|X)}+E[Z|X]^2)-E[Y|X]^2E[Z|X]^2\\&=\operatorname{var}(Y|X)\operatorname{var}(Z|X)+\operatorname{var}(Y|X)E[Z|X]^2+\operatorname{var}(Z|X)E[Y|X]^2\end{align}$$

$\endgroup$
2
  • 1
    $\begingroup$ I was so close, but thank you for the clarification! :) $\endgroup$
    – arezaie
    Sep 27 at 14:45
  • 1
    $\begingroup$ @arezaie you’re welcome $\endgroup$
    – gunes
    Sep 27 at 14:46

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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