10/07/2026 - 11:00 - LIDAM D.251 -
(Beihang University)
will give a presentation on
Abstract :
Fund ratings, as a measure of fund performance, play an increasingly important role in helping investors and financial institutions make capital allocation decisions. It is therefore crucial to develop a reliable fund ranking approach. Yang and Qiu previously proposed the expected utility-entropy (EU-E) decision model, which integrates expected utility and entropy to effectively capture both decision-makers’ subjective preferences and objective uncertainty across different states of nature. This risk measure for risky actions was further axiomatized by Luce et al. In this paper, we extend the EU-E model to a normalized expected utility-entropy-variance (EU-EV) decision model by incorporating normalized variance as an additional measure of uncertainty. We establish an alternative fund rating approach based on the EU-EV model accordingly. Furthermore, we compare the prediction ability of performance of the EU-EV model-based fund rating with that of the Lipper rating method. Using panel data regression, we employ the Sharpe ratio, Jensen’s alpha, Treynor ratio, and information ratio as performance metrics. The sample consists of 360 funds over the period from January 2010 to December 2019. The results show that the EU-EV model-based fund rating approach performs better in predicting high-star funds and is significantly superior to Lipper fund ratings in this respect. It also demonstrates stronger predictive ability for short-term and medium-term fund performance than for long-term performance.