Casey Quinn <[email protected]> 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 <[email protected]> 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/
