Recursive estimation

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

mwu888
Posts: 25
Joined: Sat Jul 09, 2011 5:49 pm

Recursive estimation

Postby mwu888 » Tue Sep 20, 2011 7:16 am

Hi,

I am struggling with doing this task in Eviews. I have an OLS equation with one independent variable (of lag 1). The sample period is from 2000M01 to 2001M12. Once I have this estimated equation, I go to View-->Stability Diagnostics-->Recursive Estimates (OLS only), I clicked on the Recursive Coefficients radio button, and checked on the box on save the output results as a series.

The output series starts on 2000M03 and ends on 2001M12.

How can I configure it so the output starts on 2002M01 and ends on 2011M08?

Thank you!

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13603
Joined: Tue Sep 16, 2008 5:38 pm

Re: Recursive estimation

Postby EViews Gareth » Tue Sep 20, 2011 7:46 am

Estimate the original equation during those dates.

mwu888
Posts: 25
Joined: Sat Jul 09, 2011 5:49 pm

Re: Recursive estimation

Postby mwu888 » Tue Sep 20, 2011 9:21 am

Gareth,

If I have a recursive regression that estimates the following:

1) Estimate 2000M01 using observations from 1990M01 to 1999M12.

2) Estimate 2000M02 using observations from 1990M01 to 2000M01.

Follwing your logic, for 1) I would set the sample period as follows: 1990M01 to 2000M01? And 2) 1990M01 to 2000M02?

Can you confirm?

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13603
Joined: Tue Sep 16, 2008 5:38 pm

Re: Recursive estimation

Postby EViews Gareth » Tue Sep 20, 2011 9:29 am

I don't understand what you're asking.

What does:
1) Estimate 2000M01 using observations from 1990M01 to 1999M12.
mean?

mwu888
Posts: 25
Joined: Sat Jul 09, 2011 5:49 pm

Re: Recursive estimation

Postby mwu888 » Tue Sep 20, 2011 12:01 pm

I meant:

Using observations from 1990M01 to 1999M12 to estimate 2000M01(forecast). Then using observations from 1990M01 to 2000M01(actual) to estimate 2000M02.

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13603
Joined: Tue Sep 16, 2008 5:38 pm

Re: Recursive estimation

Postby EViews Gareth » Tue Sep 20, 2011 12:22 pm

That's a rolling forecast.


Return to “Estimation”

Who is online

Users browsing this forum: No registered users and 2 guests