The positional MEF-TOPSIS method for the assessment of complex economic phenomena in territorial units

Abstract

In this paper, the authors propose a new methodological approach to the construction of a synthetic measure, where the objects are described by variables with strong asymmetry and extreme values (outliers). Even a single extreme value (very large or very small) of a variable for the object may significantly affect the attribution of an excessively high or low rank in the final ranking of objects. This dependence is particularly apparent when using the classical TOPSIS (Technique for Order of Preference by Similarity to Ideal Solution) method. The aim of the study is to present the application potential of the positional MEF-TOPSIS method for the assessment of the level of development of complex economic phenomena for territorial units. In the positional TOPSIS method, the application of the spatial median of Oja, which limits the impact of strong asymmetry, is proposed. In order to weaken the influence of extreme values, the Mean Excess Function (MEF) is used, by means of which it is possible to identify the limits of extreme values and establish model objects. The proposed approach is used to assess the financial self-sufficiency of Polish municipalities in 2016. The study finally compares the results of applications of positional MEF-TOPSIS and the classic and positional TOPSIS methods.