Author(s)

Hanlei Huang

Corresponding Author

Hanlei Huang

Abstract

Mutual funds represent a considerable portfolio of financial securities and provide investors an alternative way in investment. It is profound and helpful because of a lower risk than equity and derivatives and a highly qualified management. The optimal mutual fund is chosen in terms of the maximum return with a certain risk or the minimum risk with a certain return for risk-aversion investors. However, return prediction of models like dividend discount model and Residual income model is complicated due to different obstacles such as cashflow uncertainty or financial report manipulation. To solve it problem, this study will introduce some basic machine learning models like linear regression, logistic regression and decision tree. The implementation of these three algorithms involves ‘train’ and ‘test’ the data used and measures an expected return and select a suitable mutual fund for different types of investors.In addition, some model comparisons in these related models will also exhibit their performances, accuracy and complexity.

Keywords

Mutual funds;machine learning; select; return prediction; model comparisons