HAC Panel Data

For technical questions regarding estimation of single equations, systems, VARs, Factor analysis and State Space Models in EViews. General econometric questions and advice should go in the Econometric Discussions forum.

Moderators: EViews Gareth, EViews Moderator

stefanp
Posts: 63
Joined: Wed Sep 30, 2015 12:24 am

HAC Panel Data

Postby stefanp » Wed May 11, 2016 1:50 am

Is the HACestimation procedure described at http://www.eviews.com/help/helpintro.ht ... rrors.html available in panel data sets?

Equation Estimation\Panel Options has a Coef covariance drop down menu, but I cannot find the HAC method. Using Eviews 9

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13604
Joined: Tue Sep 16, 2008 5:38 pm

Re: HAC Panel Data

Postby EViews Gareth » Wed May 11, 2016 4:31 am

Since it is, econometrically, not a panel procedure, it is not available for panels.

For robust covariance options available in panels, look at the panel robust covariance section:
http://www.eviews.com/help/helpintro.ht ... 23ww187927

stefanp
Posts: 63
Joined: Wed Sep 30, 2015 12:24 am

Re: HAC Panel Data

Postby stefanp » Wed May 11, 2016 7:11 am

I think White period method works for me.

I further need to access the t-stats of the coefficients estimate. However, I end up with a message
"Square root of negative number in "GENR STATS=@TSTATS(3)"

Any ideas what causes the issue?

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13604
Joined: Tue Sep 16, 2008 5:38 pm

Re: HAC Panel Data

Postby EViews Gareth » Wed May 11, 2016 7:24 am

Would need to have the work file with the equation in question

stefanp
Posts: 63
Joined: Wed Sep 30, 2015 12:24 am

Re: HAC Panel Data

Postby stefanp » Thu May 12, 2016 12:15 am

I am afraid I cannot upload the workfile here.

However, the estimation itself does not cause any problems. Only the line

Code: Select all

STATS=@TSTATS(3)
produces the error message. Without any knowledge of the Eviews "black box" it is hard to understand that as the t-values are computed regardless of the command above. It seems like the assignment of the t-stats to the vector stats is troublesome.

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13604
Joined: Tue Sep 16, 2008 5:38 pm

Re: HAC Panel Data

Postby EViews Gareth » Thu May 12, 2016 12:19 am

What's the version/build date of your copy (Help->About EViews)?

stefanp
Posts: 63
Joined: Wed Sep 30, 2015 12:24 am

Re: HAC Panel Data

Postby stefanp » Thu May 12, 2016 2:40 am

Version 9, Nov 2 2015 build

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13604
Joined: Tue Sep 16, 2008 5:38 pm

Re: HAC Panel Data

Postby EViews Gareth » Thu May 12, 2016 4:18 am

You should update to the latest version.


Return to “Estimation”

Who is online

Users browsing this forum: No registered users and 2 guests