Can we recover the covariance matrix from its eigen-decomposition?

by hxd1011   Last Updated January 12, 2018 22:19 PM

I feel it is a very basic question, but I am still confused about it. Can we recover the covariance matrix from its eigen-decomposition?


For example, if we have a covariance matrix (of 2D Gaussian distribution) as

$$ \Sigma=\begin{bmatrix} 2 & 1 \\ 1 & 2 \end{bmatrix} $$

The eigen values are $[3,1]$, and eigen-vectors are

$$ \begin{bmatrix} \sqrt 2/2 & -\sqrt 2/2 \\ \sqrt 2/2 & \sqrt 2/2 \end{bmatrix} $$

On the other hand, can we recover $Sigma$ from its decomposition?



Related Questions




Geometric interpretation of covariance matrix

Updated May 27, 2017 04:19 AM