Purpose: The aim of this paper is to explore the identification, structuring and evaluation of the criteria for portfolio selection in an uncertain environment. It analyzes shortcomings of the traditional mean-variance framework - especially the lack of information, multidimensional risk, and multiple decision criteria, and explores how fuzzy extensions of this framework can enhance classical approaches rather than replace them.
Method: The method is based on research in the domains of multi-criteria portfolio choice, interactive deci-sion support, fuzzy portfolio evaluation and results in the development of a comparative analysis between the traditional and the fuzzy approaches. The process of portfolio selection is divided into two stages: 1) the iden-tification of efficient portfolios and 2) the ranking of efficient portfolios according to a coherent system of measures that includes the expected return as well as differentiated risk measures, diversification features and other financial metrics.
Findings: Classical models have analytical superiority and computational transparency, while fuzzy models offer a higher level of flexibility and careful consideration of uncertainty, especially in cases of market insta-bility, imperfect information and expert judgment. However, both are not mutually exclusive, oppositely they are dealing with two different levels of the same decision problem. The conclusion of the article is that the choice of methodology of optimization is not the only factor affecting the quality of the portfolio - the identi-fication, organization, weighting, and interpretation of assessment criteria are also important.