State Space Model Estimation

[George]

Dear all,

Suppose that we have a very simple state space model of the form:

@signal a=c(1)+sv1*b+[var=exp(c(2))]
@state sv1=sv1(-1)+[var=exp(c(3))]

where both "a" and "b" are observable. Suppose further that "a" appears wih a monthly frequency in the data while "b" appears with a quarterly frequency. In other words, for every 3 observations of "a" in the dataset we have 1 observation of "b". I tried to estimate this model but it seems that Eviews takes in account only those observations of "a" that appear when we also have an observation of "b" (so in fact in this case only 1/3 of the dataset seems to be used).

My questions is: can Eviews handle the estimation of this state space model taking into account ALL observations of "a" and while also properly dealing with the "missing" observations in "b"? If not, how can it be made to do so?

Thank you very much in advance for any feedback!

Regards,
George

[EViews Glenn]

I don't believe so.