DEA problems under geometrical or probability uncertainties of sample data

This paper discusses the theoretical and practical aspects of new methods for solving DEA problems under real-life geometrical uncertainty and probability uncertainty of sample data. The proposed minimax approach to solve problems with geometrical uncertainty of sample data involves an implementation of linear programming or minimax optimization, whereas the problems with probability uncertainty of sample data are solved through implementing of econometric and new stochastic optimization methods, using the stochastic frontier functions estimation. -- DEA ; sample data uncertainty ; linear programming ; minimax optimization ; stochastic optimization ; stochastic frontier functions