Page 1 of 1

Colinearities

PostPosted: Mon Jan 16, 2012 7:04 pm
by kyri82
Hello all,

When running my model I have the problem that "There are 10 eigenvalue(s) larger than 1 in modulus for 6 forward-looking variable(s)". Once I run the command "model_diagnostics(M_,options_,oo_)" I get that there is one colinear relationship in my model's equations. What does that exactly mean and where could the error be?
The full message below. Thanks!

Kyriacos

model_diagnostic: the Jacobian of the static model is singular
there is 1 colinear relationships between the variables and the equations
Colinear variables:
c1
n1
k1
y1
i1
a1
b1
mi1
lamda1
phi1
cT1
cN1
yT1
yT_I1
yN1
kT1
kN1
nT1
nN1
ksi1
pi1
zita1
tau1
c2
n2
k2
y2
i2
a2
b2
mi2
lamda2
phi2
cT2
cN2
yT2
yT_I2
yN2
kT2
kN2
nT2
nN2
ksi2
pi2
zita2
tau2
Colinear equations
Columns 1 through 11

1 2 3 4 5 6 7 8 9 10 11

Columns 12 through 22

12 13 14 15 16 17 18 19 20 21 22

Columns 23 through 32

23 24 25 26 27 28 31 32 35 36

Re: Colinearities

PostPosted: Tue Jan 17, 2012 4:49 pm
by jpfeifer
Often it means that one equation is redundant/implied by others, mostly due to Walras Law. There are quite a few posts on this issue. However, considering that you are missing 4 forward looking variables and that your model is large, I think there is something fundamentally wrong. Check again

1. If your model is correct (considering its size there may a problem be hidden here)
2. If you entered all model equations correctly
3. If the timing of the state variables is correct

Re: Colinearities

PostPosted: Tue Jan 17, 2012 6:18 pm
by kyri82
Thanks, I am working on it...

Re: Colinearities

PostPosted: Mon Sep 24, 2012 10:19 am
by 16848039
Hi guys,

How can one go about the issue of solving the problem of collinearity ?