Hello everyone,

My name is Rodrigue. Many thanks to the administrators of this new forum.

I have built a CGE model that replicates the benchmark properly. However, I observed that during the simulation, changes in some variables in a CES function are not correct. Let me consider the following equation for illustration :

Z = [a. X^p + (1-a).Y^p]^(1/p)

Where X, Y and Z are variables,

a is a share parameter and p the CES substitution parameter.

From this equation I derived the following one

Y = X.[(1-a)/a]^(1/(p-1)) with p>1

which is the central equation written in GAMS after calibrating the parameter “a” by

a = X^(p-1)/[X^(p-1)+ Y^(p-1)]

My problem relies on variable Y. The value of X being determined in another equation.

When I run the GAMS code for simulations, I observed that the percentage change in Y is exactly the same as the change in X.

I think that normally changes in X should be different from changes in Y during the computation process due to elasticity values and the share parameter. The equality holds only for a =0.5 yet in my code, all the values of a are different from 0.5.

So, can someone explain me what is wrong with my code?

I will really appreciate your help.

With regards,

Rodrigue