Error using model_steadystate Too many input arguments

Occbin
1

I am trying to replicate a paper. When I run main_exogenous_all (occbin_20140630 package needed) , I get an error that says too many input arguments. main_exogenous_all.m (5.4 KB) model.mod (8.3 KB)
model_zlb.mod (8.4 KB)

2

Did you have a look at OCCBIN in newer Dynare versions

3

Thank you Professor.

I could run the code with 4.5.6 + Original OCCBIN package.

Sorry, but I have another issue. I am trying to run the same code, but with Rotemberg Price costs (calvo- style for wages). Just eliminating the reset price and the two Xs, and replacing by the rotemberg style evolution price.
I get a problem "Unrecognized field name “ghx”.

Error in get_pq (line 3)
nvars = size(dr_.ghx,1);"

I tried with several combinations of Dynare version + updates of OCCBIN.

The error appear when I try to run the first simulation.

Really appreciate your help. Thank you. I attach the mods + matlab main file.

main_exogenous_all.m (5.4 KB)
model.mod (8.0 KB)
model_zlb.mod (8.2 KB)

4

Did you forget the param_ss.m?

5

Sorry Professor, forgot the other files and functions we need. Thank you for your time. model_ss.m (2.6 KB)
param_save.m (1.8 KB)
param_ss.m (1.9 KB)

6

The problem is that the parameters are not properly set in the mod-file. You cannot only use

load param;
7

Thank you Professor. But I still do not get it. The main model with Calvo Prices run perfectly. I just eliminated the calvo parameters and variables (calvo stickiness, x1, x2, reset price, etc) and replaced by a rotemberg style. In the parameters vector I included it. I changed the declaration order of parameters, to make it consistent. I attach again.
model_ss.m (2.6 KB)
param_save.m (1.8 KB)
param_ss.m (1.9 KB)
main_exogenous_all.m (5.4 KB)
model.mod (8.0 KB)
model_zlb.mod (8.2 KB)

The model starts running but in the first simulation it has the same error:

Unrecognized field name “ghx”.

Error in get_pq (line 3)
nvars = size(dr_.ghx,1);

Error in solve_one_constraint (line 91)
[decrulea_,decruleb_]=get_pq(oobase_.dr,Mbase_.nstatic,Mbase_.nfwrd);

8

I don’t have the original codes. But you can verify after the crash that info(1) is equal to 22 (Parameters contain NaN) and the M_.params is indeed all NaN.

9

These are the main files for the original code that runs perfectly with 4.5.6 + original occin package.

info(1) is still 22 and M.params in NaN but the code runs. I think is something about steady state/stability maybe.

main_exogenous_all.m (5.4 KB)
model_ss.m (3.0 KB)
param_save.m (1.8 KB)
param_ss.m (2.2 KB)
model.mod (8.3 KB)
model_nirp.mod (8.4 KB)
model_zlb.mod (8.4 KB)

10

On my machine, the code stops with exactly the same error. A fix should be in both mod-files:

load param;

verbatim;
for ii=1:length(M_.params)
    set_param_value(deblank(M_.param_names(ii,:)),eval(deblank(M_.param_names(ii,:))));
end
end;
11

Thank you Professor. I tried with that in both mod files and got the same problem. Tried again with two different Dynare versions + two different occbin packages. I attach the updated MOD files + main code+ functions.
In the newest Dynare version I get the error:

The class ‘dates’ is used in its own class definition as a base class, property value, or attribute
value. This is not allowed.

Error in default_option_values (line 351)
options_.dataset.firstobs = dates();

While in 4.5.6 I get the original ghx error.

model.mod (8.2 KB)
model_zlb.mod (8.2 KB)
main_exogenous_all.m (5.4 KB)
model_ss.m (2.6 KB)
param_save.m (1.8 KB)
param_ss.m (1.9 KB)

12
  1. The dates error indicates a problem with the path setting.
  2. Remove the options_.noprint to see that there now is a problem with the steady state. Is there a steady state file missing?
13
  1. Right. I had two versions, and that caused a problem with some replication files. I deleted 4.5.6 and the dates problem is solved.

My new problem now is :

Error using eval
Must be a text scalar.

Error in model.driver (line 733)
set_param_value(deblank(M_.param_names(ii,:)),eval(deblank(M_.param_names(ii,:))));

Error in dynare (line 293)
evalin(‘base’,[fname ‘.driver’]) ;

Error in solve_one_constraint (line 37)
eval(['dynare ‘,modnam_,’ noclearall nolog '])

  1. There is a function called model steady state, but I think is generated. I attach for your convenience.

So depending on the dynare version and occbin package, I still have different errors.

Thank youmodel_steadystate.m (305 Bytes)

14

The attached file runs with 4.5.7 for me (the corresponding toolkit is included)
occbin.zip (229.6 KB)

15

Thank you Professor. The file you attached is the original file with Calvo prices. I can run it too with 4.5.6 or .7. The problem is when I want to change Calvo by Rotemberg prices, as in the attached versions.

16

Again, remove the options_.noprint to see that the steady state file does not solve the model. The problem is the equation

epsip-1=epsip*exp(pm)-rotemberg*exp(infl)*exp(infl)-1+exp(Lam)*rotemberg*exp(infl(+1))*exp((infl(+1))-1);

It seems wrong. At a minimum, there is a bracket missing around (…-1)