Exploring the specific capacity of different multi criteria decision making approaches under uncertainty using data from financial markets

Abstract

Even if the MCDM methods produce statistically significant and similar rankings in a given problem, they can present the best alternatives in a different order. Random selection of the best alternative can create a complexity for the decision maker in reaching the most suitable outcome in a scenario. It is extremely challenging to oversee what the capacity or capability strengths of the more than 100 MCDM methods are, based on the results they produce. This issue is still regarded as a paradox, as there is no approved criterion to compare MCDM methods under uncertainty, in the literature. This study is aimed to determine the capacity of MCDM methods by outputs rather than inputs, unlike the previous literature. Discussions in the recent literature points out that the capacity of a MCDM method that better fits real life problems can be higher. In this respect, share returns were regarded as a reference in comparing MCDM methods objectively by financial performance of companies in this study. A multi-criteria approach that consistently produced significantly higher correlations with share returns compared to other methods has been accepted as the most appropriate MCDM method in the framework of this research. The study was conducted on 23 companies in the BIST30 index, which lists the largest companies in Borsa Istanbul. 10 MCDM methods were compared according to their significance in producing a higher relationship with share returns. As a result, PROMETHEE and FUCA methods clearly shared the first place as the most efficient compared to other methods, which are TOPSIS, GRA, S-, WSA, SAW, COPRAS, MOORA and LINMAP.