state space estimation allowing for different frequencies

[xychen1]

I have daily, monthly data in my workfile. My model in state space form is as follow,
x(t+1) = c(1) * x(t) = [var =1]
y1(t) = c(2) * x(t) + [var = exp(c(3))] ------this is in daily basis
y2(t) = c(4) * x(t) + y2(t-m) +[var = exp(c(5))] -----this is in monthly basis
I want m to automatic pick up the number of days in each month. wondering how this can b done. And also How I can extract value of x(t).
Many thanks.

[EViews Gareth]

You cannot have data of different frequencies on the same workfile page.

[xychen1]

Hi Gareth, I have one more question. Is there any way I can extract my state value x(t)

[EViews Glenn]

makestates

[xychen1]

I got an error message "Near Singular Matrix"
Not so sure what's going on...Is it possible that it is due to my model specification? Highly appreciated if someone can point out for me...
@evar var(e1) =1
@evar var(e2) = exp(c(8))
@evar var(e3) = exp(c(9))
@evar var(e4) = exp(c(10))

@evar cov(e1, e2) = 0
@evar cov(e1, e3) = 0
@evar cov(e2, e3) = 0
@evar cov(e2, e4) = 0
@evar cov(e3, e4) = 0

@state sv1 = c(1) * sv1(-1) + e1
@state u1 = c(2) * u1(-1) +e2
@state sv2 = sv1(-1)
@state sv3 = sv2(-1)
@state sv4 = sv3(-1)
@state sv5 = sv4(-1)
@state sv6 = sv5(-1)
@state sv7 = sv6(-1)

@signal de_ukx_index_c = c(3) * sv1 + u1
@signal de_pmi_c = c(4) * sv1 + de_pmi_c1m* c(5) + e3
@signal de_price_c = c(6) *sv1 + c(6) * sv2 + c(6) * sv3 + c(6) * sv4 + c(6) * sv5 +c(6) * sv7 +de_price_c1w *c(7) + e4

thanks so much