Univariate state space model + exogenous variables

Discussion of State Space and Dynamic Stochastic General Equilibrium Models
May-June
Posts: 10
Joined: Thu Dec 05, 2013 3:53 am

Univariate state space model + exogenous variables

Unread post by May-June »

Dear Tom,

I'm trying to estimate a univariate state space model including one exogenous variable in the state and one exogenous variable in the observation equation.
I got problem at the DLM estimation with the error : ## MAT2. Matrices with Dimensions 1 x 5 and 3 x 1 Involved in ~~ Operation
So, please help me to see and fix the problem.
I attached here two files:
+ The first present the Unobservable component models
+ The second show the codes Rats
And please give me some advises to deal with this model.

Many thanks
Attachments
state space 5jan2014.docx
The Unobservable Component Model
(12.31 KiB) Downloaded 765 times
TEST.RPF
The codes in Rats
(2.92 KiB) Downloaded 1019 times
TomDoan
Posts: 7814
Joined: Wed Nov 01, 2006 4:36 pm

Re: Univariate state space model + exogenous variables

Unread post by TomDoan »

The "C" matrix is entered into the model as a transpose, so CS is 3x1. You're combining that with a 1x5 CTC. BTW, CTC is a fixed matrix, so there is no reason for it to be a FRML---the only part of C that is time varying is PETROL. Either use TR(CTC) or redefine CTC as a 5x1 rather than 1x5 matrix.
May-June
Posts: 10
Joined: Thu Dec 05, 2013 3:53 am

Re: Univariate state space model + exogenous variables

Unread post by May-June »

Thank you very much. It works.
Post Reply