0
$\begingroup$

I tried to fit three variables using the non-parametric method(kernel smooth), and those distribution for each of three variables are different, so cholesky descomposition won´t work for replicate the correlation calculated by the historical data when i´m trying to make a simulation. And I need a way to simulate these variables which meets follow requirements:

  1. Each simulated variable should still follow it´s fitted non parametric distribution.

  2. Correlation should be replicated.

Anyone has an idea? I´ll apreciate.

Zhe

$\endgroup$
9
  • $\begingroup$ Please see the hits at this search. $\endgroup$
    – whuber
    Commented May 24, 2019 at 12:26
  • $\begingroup$ Hi Whuber,I just read these 2 questions posted by you which you think similar to mine, and I think no answers were about simulation for several differente non-parametric simulation. In the first one you posted,there´s no distribution limit and the second is about 2 normal distribution variable simulation. Neither will work for me, please review my question, Thanks a lot $\endgroup$
    – Zhe
    Commented May 27, 2019 at 10:56
  • $\begingroup$ Please, then, edit your post to clarify what you are looking for: that will help distinguish your problem from the apparently related ones. $\endgroup$
    – whuber
    Commented May 27, 2019 at 10:59
  • $\begingroup$ Hi, I think the need of replicating fitted non parametric distribution has been clarified to distinguish from others correlation problems. $\endgroup$
    – Zhe
    Commented May 27, 2019 at 11:19
  • 1
    $\begingroup$ Look, I´v edited my question, which clarifies 2 needs,the first is simulate non parametric distribution variables and the second is to correlate these simulated variables. I think what you just posted meets the first need and you had posted before explains how to correlate normal distribution variables,but not for correlating non parametric distribution variables $\endgroup$
    – Zhe
    Commented May 27, 2019 at 11:40

0

Browse other questions tagged or ask your own question.