Re: [Gretl-users] Variable name and future requests
Thank you for your quick answer!
Yes, I use CVS version.
Regarding
1. There is Compact data option but only to convert higher to lower
frequency not viceversa.
2. Thnx.
3. Will be nice. Thnx.
4. Ok. I'll try.
5. Thnx.
6. Ok. Thnx. There are some examples of this tests in JMulTi (also on
Sourceforge. But in Jawa). You can test group Granger causality in VECM and
you also have an option to impose zero restrictions on parameters (LR test).
From: jack <jack(a)deanovell.unian.it>
Reply-To: jack(a)deanovell.unian.it, Gretl list
<gretl-users(a)ricardo.ecn.wfu.edu>
To: Gretl list <gretl-users(a)ricardo.ecn.wfu.edu>
Subject: Re: [Gretl-users] Variable name and future requests
Date: Fri, 20 Jan 2006 10:13:56 +0100 (CET)
On Fri, 20 Jan 2006, john w wrote:
> But I also have some new suggestions for you...new features requests:
> 1. Data interpolation but from lower to high frequency data. Example
from
> quarterlly to monthly.
I don't remember if this is in 1.5.0 (I think it is), but it's definitely
in CVS.
> 2. The possibility to include exogenous variables in VAR/VECM.
Already in 1.5.0. From a script or console, you do, for example
vecm 2 1 endog1 endog2 ; exog1 exog2 --nc
In the gui you get a nice box where you can put your exogenous variables.
> 3.Var inverse roots graph but followed with textual pop-up menu. It's
> difficult to see if some value is on or out the unit circle if it is
close
> to the line.
That's a bit tricky. If you need to edit the graph, here's a workaround:
1) Right-click -> Save as Icon
2) Session->Icon View
3) Select graph->right-click-Edit code
> 4.FEVD followed by graph.
Doable. It would help if you provided a gnuplot script to show us the way
you'd like it.
> 5. More stability tests for VAR/VECM (Cusum, breakpoint, 1-step Chow
test,
> etc.)
Doable.
> 6. F-tests of zero restrictions (Wald type Granger causality) for VECM.
Doable for tests which EITHER do not involve beta OR only involve beta.
Much harder for general tests; as you probably know, Granger-causality
tests (and Wald tests in general) in cointegrated systems are not always
easy-as-pie, due to the different convergence rates among parameters.
Obligatory references:
Toda, Hiro Y & Phillips, Peter C B, 1993. "Vector Autoregressions and
Causality," Econometrica, Econometric Society, vol. 61(6), pages 1367-93.
Dolado, J.J., Lutkepohl, H. (1996), "Making wald tests work for
cointegrated VAR systems", Econometrics Reviews, Vol. 15 pp.369-86.
Riccardo `Jack' Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche
jack(a)dea.unian.it
http://www.econ.univpm.it/lucchetti
_______________________________________________
Gretl-users mailing list
Gretl-users(a)ricardo.ecn.wfu.edu
http://ricardo.ecn.wfu.edu/mailman/listinfo/gretl-users
_________________________________________________________________
The new MSN Search Toolbar now includes Desktop search!
http://toolbar.msn.co.uk/