Skip to content
Matthieu Stigler edited this page Apr 2, 2015 · 11 revisions

Frequently asked questions about tsDyn

  1. Are generalized impulse functions (GIRF) available in tsDyn?
  2. Can tsDyn estimate a TVEM with more than two variables?

1. Are generalized impulse functions (GIRF) available in tsDyn?

No, tsDyn does NOT provide generalised impulse response functions, although it does provide standard impulse response functions (IRF) for the linear VAR/VECM only, building on the functions from package vars/urca, see [irf()] (http://finzi.psych.upenn.edu/R/library/tsDyn/html/irf.html).

Several people have been asking for this functionality, and this has been discussed multiple time on the mailing list, see for example: [this thread] (http://groups.google.com/group/tsdyn/t/5c517a94a3a3ab0c). In brief, it should not be too difficult to do it with the existing functions, since tsDyn already implements a [TVECM.sim()] (http://finzi.psych.upenn.edu/R/library/tsDyn/html/TVECM.sim.html) function.

Besides of this, doing regime-specific IRF should be quite simple, by estimating a TVECM, extracting the regimes, estimating linear models on each regime and using [irf()] (http://finzi.psych.upenn.edu/R/library/tsDyn/html/irf.html). See also [this thread] (https://groups.google.com/forum/?fromgroups#!searchin/tsdyn/araujo/tsdyn/XFF6lKOjqww/F3KtFKxm9SUJ) on the mailing list.

2. Can tsDyn estimate a TVEM with more than two variables?

No, the function TVECM is restricted to two variables only. This is due to the fact that with already two variables, a two-dimensional grid search is done over the threshold and cointegrating values. If there were more than two variables, the search would have to be done over more cointegrating values, which would increase too much the dimension of the grid. People interested in using more than two variables should look at the work of [Seo (2011)] Seo, as well as [El Shagi (2011)] Shagi

Clone this wiki locally