dummy variable in state space model

[roger02]

dear moderator
i need some help here, is it possible to add a dummt variable in the observation equation ?are the state space models so flexible
to accept the dummy variable?
thank you

[EViews Glenn]

Yes.

[roger02]

many thanks

[roger02]

ok, but what about the state equation? how can i let the state variable interact with the dummy? does the state accept adding an exogenous variable?
i doubt but i would rather aske first.
thank you

[EViews Glenn]

From the manual...(p. 388-390)

"Each state equation must be linear in the one-period lag of the states."

"State equations may contain exogenous variables and unknown coefficients and me nonlinear in these elements."

"Signal equations must be linear in the contemporaneous states."

So yes, you should be able to do both...

[roger02]

Thank you glenn for your quick reply
my equation for the dummy in the signal looks like this:
@signal dependant = c(1)*crisis2005 + sv1 + sv2*independant + [var = exp(c(2))]
@state sv1 = sv1(-1) + [var = exp(c(3))]
@state sv2 = sv2(-1) + [var = exp(c(4))]
but sometimes i get this massage : "WARNING: Singular covariance - coefficients are not unique", but sometimes i get full results.
however,i tried to let the dummy interact with the signal, but couldnt do it, any ideas??
thank you

[EViews Glenn]

As with other nonlinear estimation in EViews, the state space routines take starting values from the workfile. There is no guarantee of global concavity to the likelihood functions so that optimal solutions will be starting value dependent. Moreover, it is quite possible (and as you've noted, it often happens) that you will get singularities when starting from different starting values.

[roger02]

thanx glenn
i should have paid more attention to the manual. in general, singularities happend so rarely in my previos equation.
if i asked you to rewrite the above model so that the dummy interact with the state vestor, how would you think it should be??
since its possible as you said!! i dont think i can let the dummy
thank you and merry christmas

[EViews Glenn]

Neither of your state variables has a coefficient on it so I'm not certain how interacting with a dummy variable affects anything (unless all you want to knock out a state for part of the sample). In any event, all you need to do is to multiply the state by the dummy in the signal equation.