000K  utf8
1100  $c2018
1500  eng
2050  urn:nbn:de:gbv:8-publ-22648
3000  Jin, Ding
3010  Hedtrich, Johannes
3010  Henning, Christian
4000  Applying Meta-modeling for extended CGE-modeling:Sampling techniques and potential application  [Jin, Ding]
4209  Apart from the computational time and expenses of the CGE model, the discussion of elasticity parameter estimation and various closure rules as well as the difficulty of combining the results with other analysis approaches always poses obstacles ahead of us, therefore we are motivated to apply the Bayesian model selection method and the meta-modeling technique in order to tackle these problems from a new perspective in the framework of the Senegal-CGE model and even compare the CGE models. The meta-modeling technique can be deemed as an intermediate step towards the application of Bayesian model selection method because the CGE models cannot be directly integrated into the method. The meta-modeling technique, whose three essential components are the simulation models, the meta-models and the design of experimetns, aims at generating valid and simplified approximation models of the simulation models and gives us the opportunity of combining the CGE models with the Bayesian model selection method. The purpose of this paper is to demonstrate the meta-modeling technique, test the performace of the meta-models generated by it and analyze whether various combinations of elasticity parameters affect the outputs of the CGE models which are quantified by the marginal effects. Our findings show that the produced meta-models possess a decent prediction capacity but we have not detected significant variability of the marginal effects within each unique sector.
4950  https://nbn-resolving.org/urn:nbn:de:gbv:8-publ-22648$xR$3Volltext$534
4961  https://macau.uni-kiel.de/receive/publ_mods_00002264
5051  630
5550  Bayesian model selection
5550  CGE modeling
5550  Closure Rules
5550  DOE
5550  Elasticities
5550  Meta-modeling
5550  Meta-models