0

I wanted find the following expression in closed-form

$\mathbb{E} [sgn(X_1) sgn(X_2) sgn(X_3) sgn(X_4)]$

where each $X_i$ is a gaussian random variable correlated with others. In this regard, we have to find the probability of each possible case $\mathbb{E} [sgn(X_1) sgn(X_2) sgn(X_3) sgn(X_4)] = \mathbb P(X_1>0, X_2>0, X_3>0, X_4>0) - \mathbb P(X_1<0, X_2>0, X_3>0, X_4>0) + \ldots$

here we have the closed form for $\mathbb P(X_1>0, X_2>0, X_3>0, X_4>0)$ based on this Multivariate gaussian integral over positive reals. but unfortunately not for other terms.

Please guide if you can. Thanks for your help in advance.

A. R.
  • 91
  • Why don't you mention your very recent question https://math.stackexchange.com/q/4594229/305862 ? – Jean Marie Dec 10 '22 at 18:30
  • @Jean Marie Actually, that one is the same in principle as this question. But I couldn't get an answer. I wonder if maybe a simplified question can help to get an answer. – A. R. Dec 10 '22 at 18:46

0 Answers0