Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: How to perform multivariate simulation with copula


From   "Scott Merryman" <scott.merryman@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: How to perform multivariate simulation with copula
Date   Thu, 22 May 2008 06:53:07 -0500

Casey Quinn <cq1@york.ac.uk> asked about estimating a copula using the
ml method in 2005
(http://www.stata.com/statalist/archive/2005-08/msg00929.html ; see
also his "Using copulas to measure association between ordinal
measures of health and income"
http://www.york.ac.uk/res/herc/documents/wp/07_24.pdf  ).

I don't know if he resolved the problem of finding feasible initial
values for his ml program, but you might try contacting him.

Scott


On Wed, May 14, 2008 at 7:26 AM, Matias Gutierrez <mggirault@bcra.gov.ar> wrote:
> Dear statalisters,
>
> I wonder if perhaps someone can give me a tip on how to tacke this: I have
> got three time series, say X, Y and Z which I consider are marginals from a
> multivariate joint distribution. I need to simulate 20.000 times  X, Y and Z
> being sure that those draws have embedded the dependence structure of the
> original variables. I understand that a common approach to do this is by
> using the concept of copula.
>
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index