Re: st: Nonlinear ARMAX model

[Robert A Yaffee <bob.yaffee@nyu.edu>]

Sebastien,
  That depends on what kind of nonlinear ARMAX model
you are thinking about.  If you are thinking about threshold
nonlinear autoregressive models, then you would want to be
able to test for the regime change.
  The regime could be determined by a change in the dgp from
one form of ARMAX to another.  If you are going to have a regime
switch, then the question is whether this process of regime transition
will be self-exciting or predefined by some Markov-Switching process
by which the transition probabilities have to be estimated.
  There is the question of invertbility (see Chan and tong, 2009 A note on
invertibility of nonlinear ARMA models).  Is this model supposed to be locally invertible if not globally invertible and how will you test for it.
    If you are referring to the Stata arch models, there are several
nonlinear arch models that model thresholds (tarch, atarch, tparch,
and narchk) the latter of which can model a single shift. terms
You could consult the documentation on them for more information.
    Regards,
           Bob Yaffee

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

Biosketch: http://homepages.nyu.edu/~ray1/Biosketch2009.pdf

CV:  http://homepages.nyu.edu/~ray1/vita.pdf

----- Original Message -----
From: Sebastian van Baal <s.vanbaal@arcor.de>
Date: Friday, March 5, 2010 6:12 pm
Subject: st: Nonlinear ARMAX model
To: statalist@hsphsun2.harvard.edu


> Dear Stata Users: 
>  
> Is it possible to estimate a nonlinear ARMAX model in Stata? (And if so:
> how?)
>  
> Regards 
> Sebastian
> 
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/