Computable General Equilibrium (CGE) model has become the most important tool in fields of energy, environment and economy to assess policy impact.

It is well known that the performance of the model heavily depends on the choice of parameter values, particularly elasticities. Because the number of the parameters are too large to be estimable even in models of middle size, however, most literatures have not estimated many of parameters but either just assumed or cited the values from the existing literatures even for other countries and other industries.

Recognizing these problems, the Ministry of Environment of Korea recently ordered a 7 year long research project to construct Korean CGE models in which data set will be collected and parameters will be estimated using the data set. This talk is about an result of initial research work to identify and ordain order of key parameters in a Korean CGE model. In the work, (i) a Monte Carlo simulation randomly draws exogenous parameters (X), (ii) the CGE model calculates corresponding endogenous outcomes (Y), (iii) repeat the procedure of (i) and (ii) n times to yield sample data of X and Y of size n, and (iv) finally a standard regression analysis estimates the relationship of X and Y. This approach identifies and ordains order of key parameters of the Korean CGE model.