Using DEA to estimate the importance of objectives for decision makers

In this paper we establish further connections between DEA and Multi-criteria Decision Analysis by suggesting a particular way to estimate preference weights for different objectives using DEA. We claim that the virtual multipliers obtained from a standard DEA model are not suitable to measure the preferences of a decision maker. Our suggestion takes advantage of the parallelism between DEA and the methodology proposed by Sumpsi et al. (1997) by projecting each unit on a linear combination of the elements of the pay-off matrix. Finally, we make an application of the proposed methodology to agricultural economics in a case study with Spanish data.