Appendix 22.1: Covariance Matrix Estimation

Consider the simplest case with two return series img for img and img for img where img. Truncation to S would imply that means be equal to their maximum likelihood estimates:

(22.1) equation

Here, img and img are the sample MLE for the truncated sample of size S. Similarly, the MLE covariance matrix based upon S is equal to:

(22.2) equation

where img are scalars. The parameters img and are typically ...

Get Investment Theory and Risk Management, + Website now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.