Talk:Law of total variance

Error in formula

I think there is an error in the formula:

$\operatorname {Var} [Y\mid X_{1}]=\operatorname {E} (\operatorname {Var} [Y\mid X_{1},X_{2}]\mid X_{1})+\operatorname {Var} (\operatorname {E} [Y\mid X_{1},X_{2}]\mid X_{1}).\,$

It should read:

$\operatorname {Var} [Y\mid X_{1}]=\operatorname {E} (\operatorname {Var} [Y\mid X_{1},X_{2}]\mid X_{2})+\operatorname {Var} (\operatorname {E} [Y\mid X_{1},X_{2}]\mid X_{2}).\,$ — Preceding unsigned comment added by 163.1.211.98 (talk) 16:05, 27 May 2013 (UTC)

This can't be right - the LHS of your proposed equation is measurable with respect to X_1, but the RHS is measurable with respect to X_2. This won't typically be the case. 130.88.123.107 (talk) 16:42, 3 December 2015 (UTC)

Laws of total n-th moments?

The variance is also known as the second central moment. Therefore the law of total variance could also be called the law of total second central moment. Is there also such a thing as the law of total third central moment and the law of total fourth central moment?

Multivariate case

I assume from the proof that the formula also holds for multivariate random variables. That is, if we replace Var with the Covariance-Matrix. It's not completely clear form the article right now though... --Thomasda (talk) 19:30, 9 February 2022 (UTC)