state space (copy-and-paste weird behaviour)

[maragloria]

Hi there,

I am trying to estimate a simple state model with Eviews 7.
It works. Here is the code:

@signal auqn = c(1)*cc1+ c(7)*cc2 + v1
@signal caqn = c(2)*cc1+ c(8)*cc2 + v2
@signal eurqn = c(3)*cc1+ c(9)*cc2 + v3
@signal gbqn = c(4)*cc1+ c(10)*cc2 + v4
@signal jpqn = c(5)*cc1+ c(11)*cc2 + v5
@signal nzqn = c(6)*cc1+ c(12)*cc2 + v6

@state cc1 = c(13)*cc1(-1) + c(14)*dum(-1) + [var = 1]
@state dum = cc1(-1)
@state cc2 = c(15)*cc2(-1) + [var = 1]

@state v1 = c(16)*v1(-1) + [var = exp(c(22))]
@state v2 = c(17)*v2(-1) + [var = exp(c(23))]
@state v3 = c(18)*v3(-1) + [var = exp(c(24))]
@state v4 = c(19)*v4(-1) + [var = exp(c(25))]
@state v5 = c(20)*v5(-1) + [var = exp(c(26))]
@state v6 = c(21)*v6(-1) + [var = exp(c(27))]

The problem is when I copy and paste the code into another workfile and change the variables names (their values are exactly the same), Eviews gives me the following error message:
NEAR SINGULAR MATRIX. How can that be? I just changed the variables names.

Here is the code with the changed variables names:

@signal qnau = c(1)*cc1+ c(7)*cc2 + v1
@signal qnca = c(2)*cc1+ c(8)*cc2 + v2
@signal qneu = c(3)*cc1+ c(9)*cc2 + v3
@signal qnuk = c(4)*cc1+ c(10)*cc2 + v4
@signal qnjp = c(5)*cc1+ c(11)*cc2 + v5
@signal qnnz = c(6)*cc1+ c(12)*cc2 + v6

@state cc1 = c(13)*cc1(-1) + c(14)*cc1lag(-1) + [var = 1]
@state cc1lag = cc1(-1)
@state cc2 = c(15)*cc2(-1) + [var = 1]

@state v1 = c(16)*v1(-1) + [var = exp(c(22))]
@state v2 = c(17)*v2(-1) + [var = exp(c(23))]
@state v3 = c(18)*v3(-1) + [var = exp(c(24))]
@state v4 = c(19)*v4(-1) + [var = exp(c(25))]
@state v5 = c(20)*v5(-1) + [var = exp(c(26))]
@state v6 = c(21)*v6(-1) + [var = exp(c(27))]


Thanks for your help

[startz]

One possibility is that the starting values of the coefficients a
Re different.

[maragloria]

I didn't set the starting values of the coefficients the first time.
How can I set after the copy-and-paste?

Thanks very much for your help.