1

I read many of copula articles. I almost found that the authors assumed the i.i.d observations in copula models. I just wonder if there is a reason for this assumption? I searched this site and found one question with an answer but the answer is not what I meant by this question.

Here is the link to the question:

The question is so old. I do not know how to activate it.

Maryam
  • 1,012
  • 1
  • 5
  • 20
  • How would you approach fitting a distribution to the data if the data values were interdependent? Could you tell us how you might even begin? – whuber Oct 07 '18 at 12:47
  • @whuber Thank you for your comment. Ok, is that mean we need to assume they are i.i.d to ease the estimation? – Maryam Oct 08 '18 at 05:21
  • I don't see how one could even construct a valid estimate without having more information about *how* the data fail to be independent. – whuber Oct 08 '18 at 14:39

0 Answers0