Statalist


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

Re: st: Re: No Convegence in Likelihood Estimation


From   Robert A Yaffee <[email protected]>
To   [email protected]
Subject   Re: st: Re: No Convegence in Likelihood Estimation
Date   Sat, 03 May 2008 05:02:04 -0400

Michael,
   Sometimes Arch-Garch models do not converge.   Nonlinear models do not always have
smooth response surfaces.   This is often true of multivariate Garch models.
Sometimes, you may have to try a different distribution to obtain convergence and sometimes
that may not help.. 
      - Regards,
             Robert 

Robert A. Yaffee, Ph.D.
Research Scientist
Silver School of Social Work
New York University


----- Original Message -----
From: Maarten buis <[email protected]>
Date: Friday, May 2, 2008 2:35 pm
Subject: st: Re: No Convegence in Likelihood Estimation
To: stata list <[email protected]>


> --- Michael Stobernack <[email protected]> wrote:
> > on 16 Jan 2008 you gave the following answer to a question about `No
> > convergence in Lekelihood Estimation`.
> <snip>
> > I do have similar problems. I would like to run different ARCH models
> > with a while loop in a do-File:
> >          local l = 1
> >            while `l'<=3 {
> >            capture:arch `ywahr',arima(`i',`j',`k') arch(1/`l')
> > iterate(400)
> >            local l = `l' + 1
> >           }
> > 
> > I many cases it works fine but in some cases I get the error message
> > `flat log likelihood encountered, cannot find uphill direction`.
> > Well as you mention above there might be some problems with the 
> > parameters. Since I would like to continue the while loop I use
> > capture:arch y, arch(1/1) arima(2,0,2).
> > But the computer cancels the operation and terminates.
> > Do you have any ideas how to continue the while loop in spite of an 
> 
> > error message?
> 
> First of all take a look at point 5 of the statalist FAQ
> http://www.stata.com/support/faqs/res/statalist.html
> 
> The code you sent should work, I guess that your loop also contains
> some other commands after -arch-, and that these cause you trouble. You
> can avoid that using the code below:
> 
> forvalues l = 1/3 {
>     capture:arch `ywahr',arima(`i',`j',`k') arch(1/`l') iterate(400)
>     if _rc==0 {
>         do stuff
>     }
> }
> 
> -- Maarten
> 
> -----------------------------------------
> Maarten L. Buis
> Department of Social Research Methodology
> Vrije Universiteit Amsterdam
> Boelelaan 1081
> 1081 HV Amsterdam
> The Netherlands
> 
> visiting address:
> Buitenveldertselaan 3 (Metropolitan), room Z434
> 
> +31 20 5986715
> 
> http://home.fsw.vu.nl/m.buis/
> -----------------------------------------
> 
> 
>       __________________________________________________________
> Sent from Yahoo! Mail.
> A Smarter Email http://uk.docs.yahoo.com/nowyoucan.html
> *
> *   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/
*
*   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 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index