0

I'm using a result from the dissertation of Poddar(2016, link) and he states the following in his appendix A1: We will use the well known property, stated here for completeness, of the multivariate normal distribution

enter image description here

Since this results in negative variances in my code, I'm wondering whether this is correct. No reference to the property was given by Poddar.

Kurt Z.
  • 65
  • 3
    Result (52) is standard, as discussed on this forum. Please correct the title into derivation as derivative has another meaning. – Xi'an Oct 19 '22 at 15:14
  • Also. Please correct the surname of the author, in question, in the post. – User1865345 Oct 19 '22 at 15:15
  • Consult any standard multivariate text: Anderson, Muirhead for this standard result. – User1865345 Oct 19 '22 at 15:16
  • 1
    at a guess your negative values are to do with numerical stability. computers only have finite precision (that's part of the reason they're so fast) and hence you need to structure your algorithms in a way to exploit this. https://www.johndcook.com/blog/2008/09/26/comparing-three-methods-of-computing-standard-deviation/ seems somewhat relevant, although only for scalars – Sam Mason Oct 19 '22 at 16:28
  • Please type your question as text, do not just post a photograph or screenshot (see here). When you retype the question, add the [tag:self-study] tag & read its wiki. Then tell us what you understand thus far, what you've tried & where you're stuck. We'll provide hints to help you get unstuck. Please make these changes as just posting your homework & hoping someone will do it for you is grounds for closing. – kjetil b halvorsen Oct 19 '22 at 19:09

0 Answers0