I am using IFPRI CGE (Lofgren et al., 2002) model.
1. The SAM is balanced and is checked. In caliberating it I came across with an error pointing to that weights of CPI (cwts) are not summing to unity.
Any recommendations?
2. In another case, working with already caliberated model, I wanted to simulate the case of two scenarios - on the same experiment parameter, over the same set, and under the same factor and macro-closures. The first works fine. However, in the second senario, I encountered an execution error pointing to ''negative prices'' and ''real powers''. The only difference between the second and the first scenarios is higher shocks are imposed in the second scenario.
Any recommodations for this
NB: the two questions are independent.
Regards,
Dear Amsalu,
First, you get faster answers for such CGE-GAMS questions if you sent it in GAMS-L.
Second, for the prices question, start by doing a numeraire check.
Third, for your second scenario question: you need to start tracking your error messages to know the code line where you have the problem and from there you look at the values of your calibrated parameters.
The message of zero raised to negative power is usually related to a parameter that is not correctly estimated or elasticity values that are not suitable.
Hope this helps.
Dear Prof.,
Many thanks! It was insightful. And, thanks for the GAMS-L group info too!
Regards,
1.Please try to change closure rule. Some sector may be too small to the exogenous change.
2. Negative price may be from 1. or the initial year SAM is not balanced. Run model to check consistency between left and right hand equation whether it is zero checked or not.
kitti
- Badges
- Science topic
- Similar topics
- Mathematics
- Algebra
- Matrix