Recursive estimation - changing sample?

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

tvonbrasch
Posts: 569
Joined: Fri Apr 15, 2011 5:35 am

Recursive estimation - changing sample?

Postby tvonbrasch » Fri Apr 17, 2015 8:47 am

Hi
when estimating an equation with step dummies after a specific date, the recursive estimates of the coefficients are only showed for the data points after the date of the last step dummy. For example, when looking at recursive estimates for the equation:
equation atest.ls d(a13) d(a16) @after("2010q1"),
only the dates after 2010q1 are shown. But I would like to see the recursive estimates for d(a16) beginning at an earlier date. Is there a way to do this?
Thomas

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

Re: Recursive estimation - changing sample?

Postby EViews Gareth » Fri Apr 17, 2015 9:00 am

The problem is that the dummy is identically equal to zero for all observations prior to that date. You can't estimate an equation where a variable is a constant equal to zero.

tvonbrasch
Posts: 569
Joined: Fri Apr 15, 2011 5:35 am

Re: Recursive estimation - changing sample?

Postby tvonbrasch » Sat Apr 18, 2015 9:43 am

Hi

Point taken. I was thinking of recursive estimates of the variables that do fluctuate (not the dummies). i.e. i want to check the stability of all coefficients except the dummy variables. When thinking about it, in my simple example, I guess I could estimate the two equations

equation atest.ls d(a13) d(a16) @after("2010q1")
equation atest2.ls d(a13)-c(2)*@after("2010q1") d(a16)

and then look at the recursive estimates for the second equation.
Thomas


Return to “Estimation”

Who is online

Users browsing this forum: No registered users and 2 guests