An empirical analysis to test the ability of multi-criteria methods to explain some behavioural anomalies

The analysis of the processes that lead to choices and the elements that condition them converges in decision theory Keeney and Raiffa (1993). In most cases, the theories that deal with decisions presuppose the perfect rationality of the decision-maker and therefore, the ability to identify the best choice through an essentially analytical process that can always be traced back to logical structures. In this sense, for years the theory par excellence was the theory of expected utility Von Neumann and Morgenstern (1947). In reality, however, perfect rationality does not always represent a model capable of interpreting the processes of choice Tversky and Kahneman (1974). In fact, empirical studies have shown that decision-makers implement a series of heuristics in making a choice, which causes the violation of the axioms of rationality. Re-proposing a famous Kahneman and Tversky empirical experiment based on prospects Tversky and Kahneman (1989), this work aims to study the preferences of decision-makers by relating the behavioural anomalies found and multi-criteria methods. On the obtained outputs, we test three different multi-criteria methods – ELECTRE III, PROMETHEE and TOPSIS – to determine whether they are able to explain the behaviour of the decision-makers. The results of our study show that the TOPSIS method provides a ranking that reflects the preferences expressed by the majority of individuals.