Tuesday, May 27, 2014

It's Not A Blog .......

This gem from @AcademicSay on Twitter today:

"It's not a blog. It's an independent open-access journal with a dedicated submission agreement."


© 2014, David E. Giles

3 comments:

  1. David,

    I have a question on Copula modelling.

    I would like to know whether theoretically it is possible to fit a bi-variate copula if one of the series (or both series) is (are) non-stationary. I would appeciate if you can provide me with some references on this topic.

    Thanks,
    Sriram

    ReplyDelete
    Replies
    1. Sriram - My understanding of this literature is that usually we require the series (2 or more) to be stationary. This can be checked individually and if necessary the data can be transformed to stationarity prior to applying the copula.

      There may be other ways of proceeding - I have very limited knowledge of copula theory.

      Delete
  2. Thanks David for your kind clarification.

    ReplyDelete