affine and macro affine term structure

[prodiptag]

Hi all I am trying to estimate an affine model for euro (which I plan to extend to macro-affine later on). My code is below. This works and gives sensible outputs. But I get the singular covariance warning. I have tried with a host of starting value (including OLS estimates of the states on starting date), but I get the warning no matter. Any help will be greatly appreciated

@signal y1 = sv1 + sv2*((1-exp(-c(1)))/(c(1))) + sv3*(((1-exp(-c(1)))/(c(1)))-exp(-c(1))) + [var=exp(c(2))]
@signal y2 = sv1 + sv2*((1-exp(-2*c(1)))/(c(1)*2)) + sv3*(((1-exp(-2*c(1)))/(c(1)*2))-exp(-c(1)*2)) + [var=exp(c(3))]
@signal y3 = sv1 + sv2*((1-exp(-3*c(1)))/(c(1)*3)) + sv3*(((1-exp(-3*c(1)))/(c(1)*3))-exp(-c(1)*3)) + [var=exp(c(4))]
@signal y4 = sv1 + sv2*((1-exp(-4*c(1)))/(c(1)*4)) + sv3*(((1-exp(-4*c(1)))/(c(1)*4))-exp(-c(1)*4)) + [var=exp(c(5))]
@signal y5 = sv1 + sv2*((1-exp(-5*c(1)))/(c(1)*5)) + sv3*(((1-exp(-5*c(1)))/(c(1)*5))-exp(-c(1)*5)) + [var=exp(c(6))]
@signal y6 = sv1 + sv2*((1-exp(-6*c(1)))/(c(1)*6)) + sv3*(((1-exp(-6*c(1)))/(c(1)*6))-exp(-c(1)*6)) + [var=exp(c(7))]
@signal y7 = sv1 + sv2*((1-exp(-7*c(1)))/(c(1)*7)) + sv3*(((1-exp(-7*c(1)))/(c(1)*7))-exp(-c(1)*7)) + [var=exp(c(8))]
@signal y8 = sv1 + sv2*((1-exp(-8*c(1)))/(c(1)*8)) + sv3*(((1-exp(-8*c(1)))/(c(1)*8))-exp(-c(1)*8)) + [var=exp(c(9))]
@signal y9 = sv1 + sv2*((1-exp(-9*c(1)))/(c(1)*9)) + sv3*(((1-exp(-9*c(1)))/(c(1)*9))-exp(-c(1)*9)) + [var=exp(c(10))]
@signal y10 = sv1 + sv2*((1-exp(-10*c(1)))/(c(1)*10)) + sv3*(((1-exp(-10*c(1)))/(c(1)*10))-exp(-c(1)*10)) + [var=exp(c(11))]


@state sv1=c(13) + c(14)*sv1(-1) + c(15)*sv2(-1) + c(16)*sv3(-1) + [ename=e1,var=exp(c(17))]
@state sv2 = c(18)+c(19)*sv1(-1)+c(20)*sv2(-1)+c(21)*sv3(-1)+[ename=e2,var=exp(c(22))]
@state sv3 = c(23)+c(24)*sv1(-1)+c(25)*sv2(-1)+c(26)*sv3(-1)+[ename=e3,var=exp(c(27))]

@evar cov(e1,e2) = c(30)
@evar cov(e1,e3) = c(31)
@evar cov(e2,e3) = c(32)

[prodiptag]

... which also gives all standard errors (and z-stats etc) as N/A.

BTW, I have gone through the http://forums.eviews.com/viewtopic.php?f=4&t=4808 and it was very helpful