One of the eigenvalues is close to 0/0 Not enough input arguments

Stochastic simulations
1

Hi,

I am trying to replicate a flexible-price version of two-country (advanced and emerging country) model with banking of Gertler and Karadi (2011). The closest papers to mine are Banerjee, Devereux, Lombardo (2015) and Agenor, Kharroubi, Gambacorta, Lombardo, da Silva (2017). Although it’s a flexible price model, there are prices in the code because I plan to extend the model to staggered price later on.

The attached is my Dynare file. (I tried to send the 2 matlab files but I could not because I am new user and cannot send more than 2 attachments).
What I did was, I used the matlab file to find the steady state values which I put into Dynare as initial values.

When I run my Dynare file, it seemed to be able to find the steady states. But it won’t solve and below is the message I got. Please could you assist me what went wrong with the model? I suppose it required more equations but I am not sure which variables or what kind of equations I should add?

Error using print_info (line 54)
One of the eigenvalues is close to 0/0 (the absolute value of numerator and denominator is smaller than 1e-06!
If you believe that the model has a unique solution you can try to reduce the value of qz_zero_threshold.

Error in stoch_simul (line 95)
print_info(info, options_.noprint, options_);

Error in Fed_paper_18aug (line 412)
info = stoch_simul(var_list_);

Error in dynare (line 223)
evalin(‘base’,fname) ;

Model-diagnostics suggested that:

model_diagnostics
Not enough input arguments.

Error in model_diagnostics (line 38)
endo_nbr = M.endo_nbr;

Thank you.
Fed_paper_18aug.mod (5.7 KB)

2

You need to call model_diagnostics(M_,options_,oo_) if you are using the Matlab command window.

It will say

MODEL_DIAGNOSTICS: The following endogenous variables aren’t present at the current period in the model:
lambdae
lambdac

3 Likes
3

Dear Prof. Pfeifer,

Many many thanks!! Previously I just used model_diagnostics only and it didn’t tell a lot. Calling model_diagnostics(M_,options_,oo_) is really really helpful as it tell which equations (or variables) that are conflicting.

So, the problem of my codes were in fact two: the lamdae and lambdac aren’t present at the current period and there are colinear equations. I fixed the problem and it already works!

Many thanks again!!

1 Like
4

Hi,
I also have the same problem with my code. Can you help me to check it?

Error using print_info (line 54)
One of the eigenvalues is close to 0/0 (the absolute value of numerator and
denominator is smaller than 1e-06!
If you believe that the model has a unique solution you can try to reduce the
value of qz_zero_threshold.

Error in stoch_simul (line 100)
print_info(info, options_.noprint, options_);

Error in RBC_draft (line 168)
info = stoch_simul(var_list_);

Error in dynare (line 235)
evalin(‘base’,fname) ;
RBC_draft.mod (751 Bytes)

5

Your first order conditions are strange. Please see e.g. https://github.com/JohannesPfeifer/DSGE_mod/tree/master/RBC_baseline

6

Hello,

I also have the same problem. Can someone please help me solve this issue?

Error using print_info
One of the eigenvalues is close to 0/0 (the absolute value of numerator and denominator is smaller than 1e-06! If you believe that the model has a unique solution you can try to reduce the value of qz_zero_threshold.

Error in check (line 76)
print_info(info, 0, options);

Error in modelnew (line 520)
oo_.dr.eigval = check(M_,options_,oo_);

Error in dynare (line 235)
evalin(‘base’,fname) ;

Thank you!

modelnew.mod (6.0 KB)

7

I would suggest to simplify the model. Equations like

P_o = STEADY_STATE(P_o) + epsi_o;

often result in the model being fundamentally underdetermined, because the steady state cannot endogenously be determined.