Re: TVECM code

[Matthieu Stigler]

Dear Aditi

Ccing my answer to the tsDyn mailing list.

Thanks for your kind word, I am glad tsDyn can be helpful for you!

So regarding point 1), look at the simple regime() function, which gives you a dummy variable indicating the regime. Once applied to your original dataset, this should enable you to estimate two different linear VECM and obtain regime-specific IRF. See also the recent FAQ on this point:
https://github.com/MatthieuStigler/tsDyn/wiki/FAQ

Note however that your separate models do share a common beta, which you could restrict with argument 'beta'.

Regarding the persistence profile, I am not familiar with this unfortunately.

Hope this helps,

Matthieu

2013/6/8 <aditic...@gmail.com>

Sir,

I am working on a TVECM model for moneatry measures in India. I found your working paper on Threshold Cointegration and the R code for the same very useful.
However, I had two queries:

1] Is it possible to extract each regime's model so that I can analyze them separately?

2] In order to analyze the impulse responses (IRF), is there any option to calculate the nonlinear impulse responses? Or could the individual model IRFs be calculated (using the irf() command) once I can extract the regime-wise models?
I read in a paper that Persistence Profiles were calculated instead of Generalized IRFs. Can persistence profiles be calculated in R?

Kindly look into the matter.  I shall be grateful to you if you could look into these queries.I would also like to thank you for providing the TVECM code as well as the associated threshold cointegration tests in R.


Thanking you
Yours sincerely
Aditi Chaubal

_____________________________________
Sent from http://r.789695.n4.nabble.com

[federico....@gmail.com]

Dear D. Stigler,

I'm using your amazing tsDyn package for analysing price transmission mechanism.

As I am applying a TVECM I found this thread for computing IRFs for each regime of the estimated TVECM.

Now, through the function regime() we are able to discern between the two regimes of the TVECM, but I don't know how to estimate now each regime as a linear VECM and compute the relatives IRFs, maybe somebody can help me out with that.

Secondly, I am following Lo and Zivot (2001), they developed a test for a 2-Regimes TVECM against a 3-Regimes TVECM, using a sup-LR statistics (=sup-Wald). Here again, I need some help as I do not find any package covering this kind of test and I don't have any programming skills.

Thank You in advance,

Bests

Federico

[Matthieu]

Dear Federico

For your first question, I had indicated before one should be able to run separately two VAR/VECM, I realised this is not accurate, see the exaplnationsin question 4 on the wiki:

https://github.com/MatthieuStigler/tsDyn/wiki/FAQ

For your second question, please look at: TVAR.LRtest(), and the option: test = c("1vs", "2vs3")

Le 09. 02. 16 07:58, federico....@gmail.com a écrit :

--
You received this message because you are subscribed to the Google Groups "tsdyn" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tsdyn+un...@googlegroups.com.
To post to this group, send email to ts...@googlegroups.com.
Visit this group at https://groups.google.com/group/tsdyn.
For more options, visit https://groups.google.com/d/optout.

[ahmed shafiq Joyo]

Dear sir  i am  working on price mechanism too. as seen in literature before applying TVECM model. supLM test is done to test threshold cointegration. Did you apply supLM ? your guidance is needed in this regard.

Regards