1
$\begingroup$

so I need to obtain the conditional distribution of a multivariate normal. However, I can only find it for the bivariate case:

$$(x_1|x_2=a) \sim N(\bar{\mu}, \bar{\Sigma})$$ $$\bar{\mu}= \mu_1 + \Sigma_{12}\Sigma_{22}^{-1}(a-\mu_2)$$ $$\bar{\Sigma}= \Sigma_{11}-\Sigma_{12}\Sigma_{22}^{-1}\Sigma_{21}$$

Here is the example given in more depth.

However, how would these formulas look like if I would have 3 or 4 variables instead of 2?

Edit: I would like the conditional for $(x_1| x_2=a,x_3=b)$ given I have mulitvariate distribution with 3 variables.

$\endgroup$
0

1 Answer 1

0
$\begingroup$

so the question was already answered and I just had issues with my Linear Algebra.

Here is the link:

https://stats.stackexchange.com/questions/239317/find-conditional-expectation-from-a-3-dimensional-random-vector?rq=1

$\endgroup$

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