Rolling VAR Code

[KElizabeth925]

Hello All-

I have estimated a quarterly VAR for 1948:1-2012:4. I would like to run a rolling regression for each of the VAR equations, i.e. from 1947:1-1977:1, 1947:2-1977:2...all the way through the end of the sample. I want to look at the p values of each of the coefficients and check for their stability over the sample period.

Is there a command that will allow me to perform the rolling VAR or do I need to specify this with code?

Many thanks for your help!

[EViews Gareth]

You'll have to write it yourself in a loop. There are tons of examples on the forum of running rolling estimation.

[KElizabeth925]

Thanks-found the add-in for rolling regression.
I would like to test the stability of the parameters, jointly, in a rolling regression-though only see the p values for individual coefficients when I do the roll. Have not been able to find anything on the forum relating to this. Is this something I can add in or write in program? Where is the best place to look for this.
Thanks.

[EViews Gareth]

How did you want to test for the stability of the parameters?

[KElizabeth925]

For stability of the parameters I would like to use either of the Quandt-Andrews or Chow Breakpoint...
In addition, I would really like to see a rolling Wald-test for the joint hypothesis that all lagged prices in my equation jointly=0

[EViews Gareth]

At each stage of the roll, perform the test you want, freeze the output into a table, then grab the values you're interested in from that table.

http://forums.eviews.com/viewtopic.php?f=5&t=18