Testing models of the measuring performance of mutual fund based on single and dual beta

This study aims to test some models of mutual fund performance measurement with regard to the impact of time-varying beta volatility. Testing of models based on three issues: first, the single beta testing; second, dual beta testing; and third determining which model is the most valid and robust. Tests for each models uses a two-pass regression. Testing of comparison model uses a nested model. The research samples were 30 stock mutual funds in the Indonesian capital market period January 2008-December 2012. The results research showed three finding. The first, the single beta testing indicated that these three models were not valid and were not robust. The second, dual beta testing indicated that Treynor-Mazuy model and Paramita model were valid. The third, test of robustness model showed that Paramita model was the most robust than two other models and proved that the market return variable had explanatory power as a determinant factor of mutual fund returns.