- Short tutorial about eigenvectors and eigenvalues [Link].
- It is needed to understand the relationship between multivariate Gaussians and ellipses.
- Multivariate Gaussian distributions [Link].
- The relationship between multivariate Gaussian distributions and ellipses is derived. For this purpose, the concept of isocontour is introduced.
- The expression to represent multivariate Gaussians by means of the eigenvectors and eigenvalues of the covariance matrix is presented (see the theorem and the proof in the Appendix A.2).
- The following tutorial of Linear Algebra could be useful: [Link]. It is referred to by the text in order to revise some properties of symmetric matrices.
- More on the Multivariate Gaussian, Generalized Linear Models [Link].
- An ellipse resulting from the isocontour of a multivariate Gaussian is expressed by means of the eigenvectors and eigenvalues of the covariance matrix.
