C Coefficient Vector and state space models

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

mandix
Posts: 15
Joined: Mon Jan 30, 2012 1:54 pm

C Coefficient Vector and state space models

Postby mandix » Wed Jul 18, 2012 8:46 am

Just a simple general question.
How does one change the C coefficient vector?
(I am working with state space models.)

Bonus question: say you run a state space model for the first time - you get "reasonable" results.
Then you run it again. And you get this annoying "singular matrix" message.
Was it because the *starting* C coefficient vector was *different* than the first time, and thus the starting values for the C vector was different?

Bonus question 2: is this (C coefficient vector and state space models) discussed in the User Guides?

Thanks so much to all, Manfred

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

Re: C Coefficient Vector and state space models

Postby EViews Gareth » Wed Jul 18, 2012 8:55 am

You can double click on the C vector to open it up, then change any value you want.

The C vector is updated every time any sort of estimation is done, so yes, running an estimation once then re-running it might have odd results because the starting values could be different.

Appendix B of User Guide II has information on the C(oefficient) vector and its use for starting values.

mandix
Posts: 15
Joined: Mon Jan 30, 2012 1:54 pm

Re: C Coefficient Vector and state space models

Postby mandix » Wed Jul 18, 2012 2:26 pm

Thanks Gareth.
He came, he saw, he estimated (I actually looked it up! Google Translate is a great thing....)


Return to “Estimation”

Who is online

Users browsing this forum: No registered users and 2 guests