The Transformation Properties Of Dea Method And Its Application In The Analysis Of Economic Systems

Data envelopment analysis (DEA for short), originally formulated by Charnes, is a new systerr analysis method based on the conception of the relative efficiency. DEA has become an importan analytic-al method in the analysis of economic systems in30years.But there are still many issues for further study, this dissertation constitutes the following parts around DEA method in the analysis of economic systems:Firstly, discuss the properties of translation invariance.Traditional DEA method is an efficiency assessment tool that implicitly assumes non-negativity of all inputs and outputs, this greatly limits the application in assessing the economic system. Additive model can handle zero and negative data from translation invariance, but the optimal solution of C2R or BC2model can’t keep unchanged, so chapter three discuss the change of efficiency and projection in the presence of zero and negative data.Secondly, discuss DEA and non-equilibrium complex system assessment method.Chapter four use PCA to condense the indicators, then establish a DEA model (CC-DEA) based on non-equilibrium complex system, and use it to evaluate the urban competitiveness in Shandong province.Thirdly, discuss a sample DEA method based on panel data.Chapter five use AHP to condense the indicators, then establish a sample DEA model (ED-DEA) based on panel data. This method integrates the objectivity of DEA and subjectivity of AHP.It takes advantage of the two different methods to make the results more accurate.This dissertation study from the theoretical and practical view, in theory, we give transformation of DEA model; in practice, and then establish econometrical model applied to different economic environment. Then we apply the model to evaluate the urban competitiveness of Shandong province and regional economic development level of Inner Mongolia.