EViews.com

Hi
I have to estimate a time varying AR(2)
R(t)=b0(t)+b1*R(t-1)+b2*R(t-2)+e(t)
b1(t)=b1(t-1)+u1(t)
b2(t)=b2(t-1)+u2(t)
how can I define my state space model in eviews?
Thanks

Please do a simple search in the forum before posting your questions (not to mention the users guide):
http://forums.eviews.com/viewtopic.php?f=4&t=2719

I have seen that post, but when I use this program for estimating, the information for std error-z statistics-prob is NA.

@signal y = c(1) + ar1*y(-1) + ar2*y(-2) + [var=exp(c(2))]
@state ar1 = ar1(-1) + [var=exp(c(3))]
@state ar2 = ar2(-1) + var=exp(c(4))]

the result file attached

You can find plenty of explanations and solutions to this problem if you search forum. Here is one: http://forums.eviews.com/viewtopic.php?f=4&t=2384

Dear trubador,
I searched but I couldnt fix my problem.
please help me to spesify this model correctly. and solve this my problem
thanks