Applying Meta-modeling for extended CGE-modeling:Sampling techniques and potential application

Apart from the computational time and expenses of the CGE model, the discussion of elasticity parameter estimation and various closure rules as well as the difficulty of combining the results with other analysis approaches always poses obstacles ahead of us, therefore we are motivated to apply the Bayesian model selection method and the meta-modeling technique in order to tackle these problems from a new perspective in the framework of the Senegal-CGE model and even compare the CGE models. The meta-modeling technique can be deemed as an intermediate step towards the application of Bayesian model selection method because the CGE models cannot be directly integrated into the method. The meta-modeling technique, whose three essential components are the simulation models, the meta-models and the design of experimetns, aims at generating valid and simplified approximation models of the simulation models and gives us the opportunity of combining the CGE models with the Bayesian model selection method. The purpose of this paper is to demonstrate the meta-modeling technique, test the performace of the meta-models generated by it and analyze whether various combinations of elasticity parameters affect the outputs of the CGE models which are quantified by the marginal effects. Our findings show that the produced meta-models possess a decent prediction capacity but we have not detected significant variability of the marginal effects within each unique sector.

Preview

Logo WPAP

WPAP

Rights

Use and reproduction:

No license. The provisions of the German Copyright Act (UrhG) apply.

Please note that individual components of the publication may be subject to other licensing or copyright conditions.

Cite

Citation style:
Could not load citation form.