A Problem in CGE GAMS results: ** EXIT - Other error

[Archiver]

Dear GAMS community,

My name is Herbert and currently as Phd student in energy economics fields. I have an issue regarding to the GAMS code of the CGE model I built. The issue is related to the solution of experimenting a shocked model.

So, the GAMS code for the initial state model (before shock) is working; it means I have generated the optimal solution. However, when I implementing a shock to the model, the screen result did not show an optimal solution but it showed like below:

** EXIT - Other error
Major iterations ...
Minor iterations ...
Restart...
and so on..

But before giving any shock, my model code is as follows:

** EXIT - Solution found
Major iterations ...
Minor iterations ...
Restart...
and so on..

I am pretty confused about this matter. What is actually the caused of the above error (** EXIT - Other error)? ls this error fine meaning that I could still use the changes results generated from the shock? Or should I have to ensure the Solution Found in the shocked model like it was before the shock? I have tried to put some scaling to initial values of variables but it still did not eliminate this error sign.

I hope someone could help me in this matter.

Cheers,

Herbert

--
To unsubscribe from this group and stop receiving emails from it, send an email to gamsworld+unsubscribe@googlegroups.com.
To post to this group, send email to gamsworld@googlegroups.com.
Visit this group at https://groups.google.com/group/gamsworld.
For more options, visit https://groups.google.com/d/optout.

[Archiver]

Hi Herbert

Difficult to say as I don’t have your code.

I have had models that “worked” (in my case nicely reproduced the benchmark) but failed to solve because of a stupid mistake in the equations (in one case I had a supply function with the price of the good in the denominator instead of the numerator. As benchmark prices were equal to 1, this caused no problems in the benchmark (dividing or multiplying with 1 doesn’t make a difference). However, in the counterfactual this made a huge difference…

So, if you can reproduce your benchmark (with iteration limit set to zero and all your prices and activity levels to the benchmark levels), you probably should check all your equations (a good thing to check is always the homogeneity of the equations: doubling prices shouldn’t change supply and demand. A check I now always do…).

Cheerds

Renger



From: gamsworld@googlegroups.com [mailto:gamsworld@googlegroups.com] On Behalf Of Herbert Wibert Victor Hasudungan
Sent: Monday, March 21, 2016 4:40 PM
To: gamsworld
Subject: A Problem in CGE GAMS results: ** EXIT - Other error



Dear GAMS community,



My name is Herbert and currently as Phd student in energy economics fields. I have an issue regarding to the GAMS code of the CGE model I built. The issue is related to the solution of experimenting a shocked model.



So, the GAMS code for the initial state model (before shock) is working; it means I have generated the optimal solution. However, when I implementing a shock to the model, the screen result did not show an optimal solution but it showed like below:



** EXIT - Other error

Major iterations ...

Minor iterations ...

Restart...

and so on..



But before giving any shock, my model code is as follows:



** EXIT - Solution found

Major iterations ...

Minor iterations ...

Restart...

and so on..



I am pretty confused about this matter. What is actually the caused of the above error (** EXIT - Other error)? ls this error fine meaning that I could still use the changes results generated from the shock? Or should I have to ensure the Solution Found in the shocked model like it was before the shock? I have tried to put some scaling to initial values of variables but it still did not eliminate this error sign.



I hope someone could help me in this matter.



Cheers,



Herbert

--
To unsubscribe from this group and stop receiving emails from it, send an email to gamsworld+unsubscribe@googlegroups.com.
To post to this group, send email to gamsworld@googlegroups.com.
Visit this group at https://groups.google.com/group/gamsworld.
For more options, visit https://groups.google.com/d/optout.

--
To unsubscribe from this group and stop receiving emails from it, send an email to gamsworld+unsubscribe@googlegroups.com.
To post to this group, send email to gamsworld@googlegroups.com.
Visit this group at https://groups.google.com/group/gamsworld.
For more options, visit https://groups.google.com/d/optout.

[Archiver]

Hi Renger,

Thank you very much for the quick and good responds. Hm, it looks that your advises are pretty nice. Let me try that.

However, please note that I could not attach the code since I am in the motivation to publish this work. Since this community is publicly opened. I hope you could understand this reason. I am deeply apologize for this. I thought my above question is clear enough; and indeed I saw your answers pretty good. My supervisor does not give too much support on my work (actually you can say he does not care!). Anyway thanks for this Renger, I will try to follow your advises.
I owe you some lunch mate

Herbert



On Monday, March 21, 2016 at 4:01:24 PM UTC, Renger van Nieuwkoop wrote:

Hi Herbert

Difficult to say as I don’t have your code.

I have had models that “worked” (in my case nicely reproduced the benchmark) but failed to solve because of a stupid mistake in the equations (in one case I had a supply function with the price of the good in the denominator instead of the numerator. As benchmark prices were equal to 1, this caused no problems in the benchmark (dividing or multiplying with 1 doesn’t make a difference). However, in the counterfactual this made a huge difference…

So, if you can reproduce your benchmark (with iteration limit set to zero and all your prices and activity levels to the benchmark levels), you probably should check all your equations (a good thing to check is always the homogeneity of the equations: doubling prices shouldn’t change supply and demand. A check I now always do…).

Cheerds

Renger



From: gams...@googlegroups.com [mailto:gams...@googlegroups.com] On Behalf Of Herbert Wibert Victor Hasudungan
Sent: Monday, March 21, 2016 4:40 PM
To: gamsworld
Subject: A Problem in CGE GAMS results: ** EXIT - Other error



Dear GAMS community,



My name is Herbert and currently as Phd student in energy economics fields. I have an issue regarding to the GAMS code of the CGE model I built. The issue is related to the solution of experimenting a shocked model.



So, the GAMS code for the initial state model (before shock) is working; it means I have generated the optimal solution. However, when I implementing a shock to the model, the screen result did not show an optimal solution but it showed like below:



** EXIT - Other error

Major iterations ...

Minor iterations ...

Restart...

and so on..



But before giving any shock, my model code is as follows:



** EXIT - Solution found

Major iterations ...

Minor iterations ...

Restart...

and so on..



I am pretty confused about this matter. What is actually the caused of the above error (** EXIT - Other error)? ls this error fine meaning that I could still use the changes results generated from the shock? Or should I have to ensure the Solution Found in the shocked model like it was before the shock? I have tried to put some scaling to initial values of variables but it still did not eliminate this error sign.



I hope someone could help me in this matter.



Cheers,



Herbert

--
To unsubscribe from this group and stop receiving emails from it, send an email to gamsworld+...@googlegroups.com.
To post to this group, send email to gams...@googlegroups.com.
Visit this group at https://groups.google.com/group/gamsworld.
For more options, visit https://groups.google.com/d/optout.

--
To unsubscribe from this group and stop receiving emails from it, send an email to gamsworld+unsubscribe@googlegroups.com.
To post to this group, send email to gamsworld@googlegroups.com.
Visit this group at https://groups.google.com/group/gamsworld.
For more options, visit https://groups.google.com/d/optout.

[Archiver]

Thanks for this Renger, it helps alot. Cheers, Herbert

On Monday, March 21, 2016 at 4:01:24 PM UTC, Renger van Nieuwkoop wrote:

Hi Herbert

Difficult to say as I don’t have your code.

I have had models that “worked” (in my case nicely reproduced the benchmark) but failed to solve because of a stupid mistake in the equations (in one case I had a supply function with the price of the good in the denominator instead of the numerator. As benchmark prices were equal to 1, this caused no problems in the benchmark (dividing or multiplying with 1 doesn’t make a difference). However, in the counterfactual this made a huge difference…

So, if you can reproduce your benchmark (with iteration limit set to zero and all your prices and activity levels to the benchmark levels), you probably should check all your equations (a good thing to check is always the homogeneity of the equations: doubling prices shouldn’t change supply and demand. A check I now always do…).

Cheerds

Renger



From: gams...@googlegroups.com [mailto:gams...@googlegroups.com] On Behalf Of Herbert Wibert Victor Hasudungan
Sent: Monday, March 21, 2016 4:40 PM
To: gamsworld
Subject: A Problem in CGE GAMS results: ** EXIT - Other error



Dear GAMS community,



My name is Herbert and currently as Phd student in energy economics fields. I have an issue regarding to the GAMS code of the CGE model I built. The issue is related to the solution of experimenting a shocked model.



So, the GAMS code for the initial state model (before shock) is working; it means I have generated the optimal solution. However, when I implementing a shock to the model, the screen result did not show an optimal solution but it showed like below:



** EXIT - Other error

Major iterations ...

Minor iterations ...

Restart...

and so on..



But before giving any shock, my model code is as follows:



** EXIT - Solution found

Major iterations ...

Minor iterations ...

Restart...

and so on..



I am pretty confused about this matter. What is actually the caused of the above error (** EXIT - Other error)? ls this error fine meaning that I could still use the changes results generated from the shock? Or should I have to ensure the Solution Found in the shocked model like it was before the shock? I have tried to put some scaling to initial values of variables but it still did not eliminate this error sign.



I hope someone could help me in this matter.



Cheers,



Herbert

--
To unsubscribe from this group and stop receiving emails from it, send an email to gamsworld+...@googlegroups.com.
To post to this group, send email to gams...@googlegroups.com.
Visit this group at https://groups.google.com/group/gamsworld.
For more options, visit https://groups.google.com/d/optout.

--
To unsubscribe from this group and stop receiving emails from it, send an email to gamsworld+unsubscribe@googlegroups.com.
To post to this group, send email to gamsworld@googlegroups.com.
Visit this group at https://groups.google.com/group/gamsworld.
For more options, visit https://groups.google.com/d/optout.