3
$\begingroup$

Let $\begin{bmatrix} Y_1 \\ Y_2 \\ Y_3 \end{bmatrix} \sim MVN\left( \begin{bmatrix} \mu_1 \\ \mu_2 \\ \mu_3 \end{bmatrix} , \begin{bmatrix} \Sigma_{11} & \Sigma_{12} & \Sigma_{13} \\ \Sigma_{21} & \Sigma_{22} & \Sigma_{23} \\ \Sigma_{31} & \Sigma_{32} & \Sigma_{33} \end{bmatrix}\right)$ where MVN denotes the multivariate normal distribution. I wish to calculate the conditional expectation $E[Y_3 | Y_1 = y_1]$, is it true then I can just ignore $Y_2$ basically and consider:

$$\begin{bmatrix} Y_1 \\ Y_3 \end{bmatrix} \sim MVN\left( \begin{bmatrix} \mu_1 \\ \mu_3 \end{bmatrix} , \begin{bmatrix} \Sigma_{11} & \Sigma_{13} \\ \Sigma_{31} & \Sigma_{33} \end{bmatrix}\right).$$ That is, $(Y_1, Y_3)$ are jointly bivariate normally distributed as above?

$\endgroup$

1 Answer 1

1
$\begingroup$

Yes, you are correct.

If $X$ is a MVN random vector, then every component of $X$ is univariate Normal and every subvector of $X$ is also MVN, with appropriate partition of the the mean and covariance martrix. Further all conditional distributions are also normal.

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .