I have question that I believe is relatively easy, but a clear answer
has eluded me.
I have a log-differenced time series variable, "x", with some
autoregressive conditional heteroskedasticity. I would like to filter
the variable of these effects, so that I can use it elsewhere (in a
separate VAR model). I've estimated the GARCH (1,1) coefficients (see
below), but I'm unsure how to proceed. I've tried a number of options,
but I'm not confident which is correct.
. arch x, arch(1) garch(1)
...omitted...
ARCH family regression
Sample: 1960q1 to 2005q4 Number of obs =
184
Wald chi2(.) =
.
Log likelihood = -761.3947 Prob > chi2 =
.
------------------------------------------------------------------------
------
| OPG
x | Coef. Std. Err. z P>|z| [95% Conf.
Interval]
-------------+----------------------------------------------------------
------
x |
_cons | 4.141107 .882965 4.69 0.000 2.410528
5.871687
-------------+----------------------------------------------------------
------
ARCH |
arch |
L1. | .6885315 .165625 4.16 0.000 .3639126
1.01315
garch |
L1. | .395975 .05436 7.28 0.000 .2894314
.5025186
_cons | 12.62181 8.695377 1.45 0.147 -4.420812
29.66444
------------------------------------------------------------------------
------
I would greatly appreciate any guidance on how to obtain the "clean"
version of my variable.
Thanks.
Jeffrey W Ladewig
Assistant Professor
University of Connecticut
Department of Political Science
341 Mansfield Road, U-1024
Storrs, Connecticut 06269-1024
(860)486-3747
*
* 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/
