# Model runs but i get the problem"colinear relationships "

Hello everyone!

I begin a model with two types of households and two sectors. After solving the steady state, I run the model, I find it can figure out the impulse response. However, the Monetary policy shocks appear to be contrary to normal, with positive shocks causing output to increase.

So I use the model_diagnostics(M_,options_,oo_), and it indicate:

“MODEL_DIAGNOSTICS: The Jacobian of the static model is singular
MODEL_DIAGNOSTICS: there is 1 colinear relationships between the variables and the equations
Colinear variables:
ch
ds
bb
bk
nbk
yk
hs
hb
k
r
eta_h
lamda_b
lamda_h
lamda_k
Pi
Colinear equations

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

21 22 23 24 25 26 27 30 31 33 34 36
MODEL_DIAGNOSTICS: The presence of a singularity problem typically indicates that there is one
MODEL_DIAGNOSTICS: redundant equation entered in the model block, while another non-redundant equation
MODEL_DIAGNOSTICS: is missing. The problem often derives from Walras Law.”

Actually, I check my equations, and I make sure I omit one budget constraint condition for the Walras Law, so the number of vars is equal to equations, and not a redundant equation.

What also makes me confused is when I increase the parameter of interest rate matching
targeting coefficient of the inflation target in the Taylor Rule， then it can not run with the error:

How can I solve this problem? Could anyone give any suggestions about it?

Here is my code.
thy5.mod (7.5 KB)

Thank you very much in advance!

With a model of that size, usually the only option is to radically simplify the model to find out where the issue arises.

Dr.jpfeifer, thank you for the suggestion! I have run a simple model that doesn’t consider the consumption inertia. And the model runs well. In order to have a hump-shape impulse response, I consider the consumption inertia, just introducing some Lagrange operators. Then the model meets these errors, what can I do next ?

Then you need to find out why that creates an issue. Sometimes it’s just a numerical thing with steady state Lagrange multipliers having to large a value in steady state so that numerical overflow happens. In that case, you can often ignore the singularity warning.

Thank you very much, Dr.jpfeifer! I still have other two problems:

1. I find I can not set the parameter of interest rate matching
targeting coefficient of the inflation target in the Taylor Rul (rho_PI) over a number (approximately 1).
``````r/R = (r(-1)/R)^rho_r*(Pi/PI)^((1-rho_r)*rho_PI)*(y/Y)^((1-rho_r)*rho_Y)*exp(em)
``````
1. I find the positive shock of em causes output (y) consumption (c) to increase, which contradicts the normal impulse response.
Dr.jpfeifer, what may cause these problems and what can I do for it?

That usually suggests a fundamental problem in your model, typically a timing error but sometimes just a matter with the parameterization.

My personal experience about the model collinearity problem goes back to the model structure for example modeling of Taxes , Bonds and so on.

When you have timming problem in your model in many cases you encounter with Blanchard-Kahn failure in Dynare results.

1 Like

Thank you, Dr.jpfeifer! I will check the model again.

Thank you! I will check the structure again.