given $\begin{bmatrix}X\\Y\end{bmatrix}\sim N\left(\begin{bmatrix}\mu_X\\\mu_Y\end{bmatrix};\begin{bmatrix}\Sigma_{XX}&\Sigma_{XY}\\\Sigma_{YX}&\Sigma_{YY}\end{bmatrix}\right)$ How can I prove that $\Sigma_{YX}-\Sigma_{YX}\Sigma_{XX}^{-}\Sigma_{XX}=0$, where $\Sigma_{XX}^{-}$ is the reflexive g-inverse, i.e. $\Sigma_{XX}\Sigma_{XX}^{-}\Sigma_{XX}=\Sigma_{XX}$.

The bigger picture is that I am trying to find en expression for the conditional distribution $Y|X=x$ when $\Sigma_{XX}$ is singular. My approach is to prove that $X$ and $Y-\Sigma_{YX}\Sigma_{XX}^{-}X$ are independent. Once this is proved the rest is trivial. However, I got stuck with the problem above.


Your Answer

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

Browse other questions tagged or ask your own question.