I tried tseries::garch() and was getting a lot of False Convergence so 
tried fGarch::grachFit.  

Looking a bit further I find that from fGarch::garchFit I get

 ..@ fit        :List of 17
 <snip>
  .. ..$ convergence: int 1
  .. ..$ message    : chr "singular convergence (7)"

for any and all fits.  I have tried various real and lots of simulated 
data including a wide range of garch(1,1) simulations.  The @fitted 
values are all the same; the @h.t and @sigma.t seem reasonable. Am I 
interpreting  the @fitted incorrectly and is the convergence error 
perhaps bogus it would seem in at least some of my many tries should 
have converged.

Ron

Liviu Andronic wrote:
> Hello,
>
> On 7/1/09, Ron Burns <rrbu...@cox.net> wrote:
>   
>>  In trying to fit garch models in above environment. I am getting
>>  "reasonable" fitted coefficients, but the fitobj...@fitted are all the
>>  same. This is true even for the help page example:
>>
>>     
> There is a better chance of getting good answers if asking finance
> related questions on r-sig-finance. Concerning your question, have you
> tried fitting the same examples with tseries::garch() or
> rgarch::ugarchfit() [1]?
> Liviu
>
> [1] http://rgarch.r-forge.r-project.org/index.html
>
>   


-- 
R. R. Burns
Physicist (Retired)
Oceanside, CA


        [[alternative HTML version deleted]]

______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Reply via email to