This paper focuses on practical application of the ideas of prospect theory in order to propose a new methodology for mutual funds ranking. Method used in the research is regression analysis of purchases and sales of investment units of large Russian mutual funds. Analyzing these indicators authors designed econometric model and assessed the loss aversion coefficient inherent to Russian investors. Henceforth, this allowed to concretize Kahneman-Tversky utility function and apply it to portfolio assessment. Verification of the model displayed that behavior of investors is comparable to the postulates of prospect theory. Based on the findings, new ranking principle was suggested, which is fundamentally distinct from the traditional ones (Sharpe-ratio, Morningstar Risk-Adjusted Return Approach). Thus, the research results may be valuable for fund governance as new ratings expand investment strategies that are admissible to unqualified investors.