Hello. Thank you for all the help here, it has been really helpfull.
I am making a Kalman Filter model to get an unonbserved factor from two series (as Engle and Watson).
The thing is I have a doubt with the representation, for example:
KF.append @state FS_Unem = FS_Unem(-1) + [var=exp(c(1))]
KF.append @state FS_GDP = FS_GDP(-1) + [var=exp(c(2))]
KF.append @state FC = FC(-1) + [var=exp(c(3))]
KF.append @signal Unem = FC + FS_Des + [var = exp(c(4))]
KF.append @signal GDP = FC + FS_CtPIB + [var = exp(c(6))]
Here I have two signals for variables Unem and GDP, and the states for the specific factors of each series and a comon factor to them.
The thing is I don´t know some things about this:
-If with this set out Eviews will automatically set the initial conditions for the unknown factors (FS_Unem, FS_GDP and FC) to estimate them.
-Also I am not sure if I have to add the coefficients with the variables or if they will be given by default (for example, instead of writing FC=FC+FS_Des +[var = exp(c(6))] I need to put FC=c(6)*FC+c(7)*FS_Des +[var = exp(c(6))]).
Thanks a lot.
Regards.
State Space
Moderators: EViews Gareth, EViews Jason, EViews Moderator, EViews Matt
-
- EViews Developer
- Posts: 2672
- Joined: Wed Oct 15, 2008 9:17 am
Re: State Space
EViews will always set initial conditions unless you provide them explicitly. We'll solve for the ergodic if possible, otherwise we'll initialize to diffuse as described in the manual.
If you have any coefficients in the observed or state equations, you'll have to provide them explicitly.
If you have any coefficients in the observed or state equations, you'll have to provide them explicitly.
Re: State Space
Ok, thank you!
I will try this.
Regards.
I will try this.
Regards.
Who is online
Users browsing this forum: No registered users and 20 guests