This project is read-only.

Exogenous regressors in a model

Topics: Models
Mar 27, 2014 at 7:40 PM
Hi folks, I was wondering if there is a way to implement a model of the following form:

Transition Variables
x1

Measurement Variables
y1, y2

Parameters:
A, B

Transition block:
x1t = A * x1t-1 + B y2t + et;

Measurement block:
y1t = C * x1t + vt;
y2t = y2t;

IRIS throws an error because y2 is not a transition variable.

I tried (!exogenous_variables) without success as these variables only work on the !dtrends block. What is the most appropriate way to set up this model?

Thanks!

Hector
Apr 2, 2014 at 8:46 AM
Edited Apr 2, 2014 at 9:17 AM
Since y2 is entering your state equation so you need to specify a process for it. Needs to look something like this.
!transition_variables
x1,y2s

!measurement_variables
y1,y2

!parameters
a,b

!transition_equations
x1 = a * x1{-1} + b * y2s + e ;
y2s = y2s_shk ;

!measurement_equations
y1 = c * x1 + v ;
y2 = y2s ;

!transition_shocks
et,y2s_shk,