Threshold Structural VAR

For questions about EViews Add-ins available from the EViews Add-ins webpage. Note each add-in available on our webpage will have its own individual thread.

Moderators: EViews Gareth, EViews Moderator, EViews Esther

nasa
Posts: 39
Joined: Tue Aug 02, 2016 12:41 am

Re: Threshold Structural VAR

Postby nasa » Sun Aug 20, 2017 12:47 am

Hi,

Here is my work file and I think based on the formula, the following is the result.Is that right,please?
The endogenous variables are: dgdp,dspending,dtax and the threshold variable and response variable is dgdp and the lag length for the VAR is 2.
Degrees of freedom=N-1.
b = 0.15*(2014q2-1960q1+1)+3-217
= 0.15*(218+1)+3-217
=-181.15
thsvar workfile.WF1
(151.16 KiB) Downloaded 13 times


Best.

dakila
Posts: 264
Joined: Tue Nov 24, 2015 4:57 pm

Re: Threshold Structural VAR

Postby dakila » Sun Aug 20, 2017 2:04 am

I don't think so. The degrees of freedom is just for adjustment of the sample size. It should not be big number. It may be between 0 to 5.

nasa
Posts: 39
Joined: Tue Aug 02, 2016 12:41 am

Re: Threshold Structural VAR

Postby nasa » Sun Aug 20, 2017 2:14 am

ok,

may be the formula is wrong or what is my wrong here is then?

plus can i get the impulses result in table form,please?

Best.

nasa
Posts: 39
Joined: Tue Aug 02, 2016 12:41 am

Re: Threshold Structural VAR

Postby nasa » Fri Sep 08, 2017 12:26 am

Hi,

I think that the thsvar will be very useful add in for econometricians if it can include confidence intervals based on Monte Carlo simulations for the GIRFs. It could help them to better interpret the results.

Best

dakila
Posts: 264
Joined: Tue Nov 24, 2015 4:57 pm

Re: Threshold Structural VAR

Postby dakila » Mon Sep 11, 2017 3:40 pm

thanks for the comment

EdgarM
Posts: 4
Joined: Fri Jul 28, 2017 12:00 am

Re: Threshold Structural VAR

Postby EdgarM » Thu Oct 19, 2017 7:10 am

Hello, thanks in advance for the add-in it was really usefull.

I have a question: When i estimate the TVAR with 3 dependent variables, i got 6 outputs instead of only one. I was replied that if i gather all the information in one matrix it was the same, but when i try to analize for example, serial correlation, the results are different for each regression.
How can i merge all of them in one test to unify the results? (or where can i study it, i've been searching in the web for a few days now without results).
Best regards,
Edgar

dakila
Posts: 264
Joined: Tue Nov 24, 2015 4:57 pm

Re: Threshold Structural VAR

Postby dakila » Fri Oct 20, 2017 5:16 pm

Hello Edgar,

You can do it easily. For example, first estimate VAR model using the following code or the dialog box:

Code: Select all

smpl 1960q1 1997q3 if d1=1
var var01.ls 1 4 d1gdp d1pgdp fyff cpbill1

Then use eviews VAR model test.

EdgarM
Posts: 4
Joined: Fri Jul 28, 2017 12:00 am

Re: Threshold Structural VAR

Postby EdgarM » Tue Oct 24, 2017 1:09 pm

Thank you very much Dakila!

banhbengconuong
Posts: 15
Joined: Mon Apr 25, 2011 8:31 pm

Re: Threshold Structural VAR

Postby banhbengconuong » Thu Nov 16, 2017 6:18 am

Dear Dakila,

Thanks so much for making the add-in. I would like to ask whether we can have 2 thresholds for a variable?

Bests,

dakila
Posts: 264
Joined: Tue Nov 24, 2015 4:57 pm

Re: Threshold Structural VAR

Postby dakila » Thu Nov 16, 2017 7:31 am

No


Return to “Add-in Support”

Who is online

Users browsing this forum: No registered users and 1 guest