Given the equation: $V_k = M v_k - A M v_{k-1}$ with $ M = [C^T C]^{-1} C^T,$

where capital letters represent matrices. A paper that I am reading ( see equation (27) here) states that you can compute the covariance of $V_k$ when $C, A$ and $cov(v_k)$ is known.
I don't see how, though. Is the covariance linear?

  • $\begingroup$ Write $ symbols to the left and right side of the LaTeX. $\endgroup$ – Peter Foreman Feb 9 at 21:28

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.