Coeficient vector C: sepcification

For questions regarding programming in the EViews programming language.

Moderators: EViews Gareth, EViews Moderator, EViews Jason, EViews Matt

frhes
Posts: 2
Joined: Tue Oct 08, 2013 12:48 am

Coeficient vector C: sepcification

Postby frhes » Tue Oct 08, 2013 12:55 am

Is there any information available how the dimension of the coefficient vector is specified by Eviews internally and how it could be modified? I am confronted with a 750 x 1 coefficient vector in a state space model which prevents yhe estimation of the Kalman filter due to insufficient observations. The dimension of this coefficient vector does not respond to the specifications of my state-space model.

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

Re: Coeficient vector C: sepcification

Postby EViews Gareth » Tue Oct 08, 2013 7:47 am

It is a 750 row vector. What more do you need to know?

frhes
Posts: 2
Joined: Tue Oct 08, 2013 12:48 am

Re: Coeficient vector C: sepcification

Postby frhes » Tue Oct 08, 2013 8:20 am

Indeed it is a 750 row vector, but how did EVIEWS choose the number of rows? And why does this number not react to respecifications of the state space model which change the number of coefficients?

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

Re: Coeficient vector C: sepcification

Postby EViews Gareth » Tue Oct 08, 2013 8:30 am

It is always 750. EViews doesn't choose the size. It is always 750.

startz
Non-normality and collinearity are NOT problems!
Posts: 3798
Joined: Wed Sep 17, 2008 2:25 pm

Re: Coeficient vector C: sepcification

Postby startz » Tue Oct 08, 2013 9:36 am

EViews sets aside 750 spaces in C, but doesn't use the ones that don't show up in the model.


Return to “Programming”

Who is online

Users browsing this forum: No registered users and 1 guest