Meta Response Surface Design for General and Partial Equilibrium Models

Due to the complexity of general and partialequilibrium models, conventional sensitivity analysis,qualitative reviews or literature-based meta-analyses donot allow for detailed assessments of the role ofindividual parameters and policy shocks across differentmodels. Therefore, the partial equilibrium model“GSIM” and a single country CGE are employed togenerate synthetic scenarios based on randomlyspecified combinations of base data, elasticities andtariff changes selected from previously specified,plausible ranges. The synthetic meta-data has theadvantage that the values of explanatory variables aremeasured exactly. This makes it possible to explorecomplex issues of functional form and interactionbetween variables in the estimation of the responsesurface of each model as well as for a joint responsesurface of both models. The results indicate that firstandsecond-order polynomials provide sufficientapproximations of the model responses, and especiallyfor the CGE model, interaction terms of elasticities withpolicy variables play an important role. Furthermore,simultaneous estimation of a response surface ofscenarios from both models proves to be feasible andenables quantitative comparisons of different modeloutput, e.g. welfare measures.