Dynare forums

Hello,

I'm trying to simulate this model and the messages below is showing up, I have not found errors.

Please, help me.

(model follows in attachment)

Octave-3.6.4:
error: linsolve: matrix singular to machine precision
terminate called after throwing an instance of 'octave_execution_exception'

This application has requested the Runtime to terminate it in an unusaul way.


Matlab 7.0:


There are 12 eigenvalue(s) larger than 1 in modulus
for 12 forward-looking variable(s)

The rank condition ISN'T verified!


STEADY-STATE RESULTS:

Y 0
C 0
I 0
K 0
K_C 0
K_I 0
L 0
L_C 0
L_I 0
W 0
R 0
R_n 0
R_f 0
PI 0
PI_C 0
PI_I 0
P 0
P_C 0
P_I 0
U 0
Q 0
Xi 0
N 0
A_C 0
A_I 0
S_C 0
S_L 0
Warning: Log of zero.
> In dyn_first_order_solver at 268
In stochastic_solvers at 217
In resol at 113
In stoch_simul at 80
In Modelo2b at 304
In dynare at 162
??? Error using ==> print_info
Blanchard Kahn conditions are not satisfied: indeterminacy due to rank failure

Error in ==> stoch_simul at 85
print_info(info, options_.noprint, options_);

Error in ==> Modelo2b at 304
info = stoch_simul(var_list_);

Error in ==> dynare at 162
evalin('base',fname) ;

Johannes,

Please, I need your help.

best regards

Celso

It's hard to tell, but your inflation steady states look strange (zero?). It seems as if you are confusing net and gross inflation. The former is zero and the latter is 1. Look at your Taylor Rule. With pi_ss=0, inflation completely drops out.

Johannes,

Thanks for your help.

I was uncomfortable with the taylor rule without inflation (because inflation in the steady state is zero - in the model). But if the rate of inflation is the price change between periods (PI = P/P(-1) -1), and if the price is fixed at steady state, would not be zero inflation in the steady state?
What is the matter?

best regards,

Celso

Dear Celso,

there is a difference between gross inflation (P_t/P_{t-1}) and net inflation (P_t/P_{t-1}-1) (and correspondingly for gross and net interest rates). It is always net inflation that is set to 0, not gross inflation. Your Taylor rule in unlinearized form features gross interest rates and gross inflation rates. The log-linearization of gross interest and inflation rates is approximately the net interest and inflation rate. However, the steady state around which you are approximating still must refer to the gross rates. This is your mistake. You take Pi_ss, the gross rate to be 0 although it should be 1. The "A Guide to Specifying Observation Equations for the Estimation of DSGE Models" at https://sites.google.com/site/pfeiferecon/Pfeifer_2013_Observation_Equations.pdf shows this in more detail.

Dear Johannes,

I saw the problem of rule taylor in the my model. I made some changes, now the Blanchard-Kahn conditions are not satisfied.
There are 6 eigenvalue (s) larger than 1 in modulus
5 is forward-looking variable (s)

What do I do?

Thanks very much for your help.

Best regards,

Celso