Problem with example RBC_Est - linking data to variables


I apologize in advance for a probably easy question, but I’m very new to Dynare and been struggling with this for a while.

I try to go through example code RBC_Est, but with slight difference, that is instead of simulated data I try to use HP filtered GDP. The data is stored in .xls file (one column named the same as in varobs). I get some errors I have troubles solving out. What I’m especially worried about is the line:
“You did not declare endogenous variables after the estimation command.”
I’ve been experimenting with inputting the data through other formats, but I repeatedly get the above message. I’ve even tried no to give any datafile in the workspace at all and check whether this message occurs and it did - that is why I think there is something wrong with reading the data.

I would be extremely thankful for any advice. Thanks in advance.


Below the error message and files used in attachments.

dynare RBC_Est

Configuring Dynare …
[mex] Generalized QZ.
[mex] Sylvester equation solution.
[mex] Kronecker products.
[mex] Sparse kronecker products.

Starting Dynare …
Starting preprocessing of the model file …
9 equation(s) found
Processing derivation …
Processing Order 1… done
Processing Order 2… done
Processing outputs …
Preprocessing completed.
Starting Matlab computing …

You did not declare endogenous variables after the estimation command.

You did not declare endogenous variables after the estimation command.
??? Error using ==> print
Error using ==> graphics\private\ghostscript
Problem calling GhostScript. System returned error

Error in ==> plot_priors at 68
eval('print -dpdf ’ M_.fname ‘_Priors’ int2str(1)]);

Error in ==> dynare_estimation_1 at 92

Error in ==> dynare_estimation at 62

Error in ==> RBC_Est at 130

Error in ==> dynare at 102
evalin(‘base’,fname) ;

RBC_Est.mod (779 Bytes)
HPcycle.xls (18.5 KB)

“You did not declare endogenous variables after the estimation command.” isn’t an error but a warning. Depending what kind of statistics you request at the estimation stage, it may generate a big quantity of results. It is better to specify which variables you are interested in at the end of the estimation command.



Thanks Michel :slight_smile: - you really helped me out with this line

I was also wondering what other error messages mean. The ghostscript error is, I suppose, something internal to my matlab, but maybe you have come across those other error messages?

I really appreciate your help. Dynare team rulez :slight_smile:.

best regards,

You are right the ghostscript error is Matlab related. I remember having encountered such problems a couple of years ago installing Matlab in a Linux system.
The easiest solution short of solving the ghoscript problem on your installation is to comment out line 68 in plot_priors.m It will only prevent Dynare to saving prior graphs in PDF format. You will certainly have to comment out additional similar lines elsewhere in Dynare code