MUTUAL FUND PERFORMANCE PREDICTION MODEL

The purpose of this research is to create a predictive model for mutual fund ratings and returns. This model can aid financial advisors and investors when making decisions regarding mutual fund investments. We obtained a... [ view full abstract ]

The purpose of this research is to create a predictive model for mutual fund ratings and returns. This model can aid financial advisors and investors when making decisions regarding mutual fund investments. We obtained a systematic sample of 500 mutual funds from a larger sample of 11,000 mutual funds. Sixty funds were left out of the data analysis in order to test the predictive accuracy of the model. The larger sample was acquired from Wells Fargo advisers. All sampled funds were no load mutual funds. The sampled mutual funds had a mix of large and small market capitalization. For this study, we used the following variables as independent variables: Fund category (Morningstar), return (factor calculated based on one-year return percent, three year return percent, five year return percent, ten year return percent, since inception return percent), Fund age, expenses, alpha, beta, standard deviation, Sharpe ratio, and net assets of funds. The risk and the Morningstar Rating are used as dependent variables to determine the contribution of the each of the aforementioned variable to fund rating and the associated risk. A regression analysis showed a moderate relationship between independent and dependent variables. Future research will include a more comprehensive factor analysis to reduce the number of independent variables and a cluster analysis of a larger sample of mutual funds.