An Expected Return Model

摘要

We have developed a new expected return model for an investment software package, called PORSEL PORSEL (PORtfolio SELection system), is an expert system based on fuzzy logic which uses a small set of rules to select stocks. This model improves the PORSEL performance by incorporating a better method of constructing portfolios. The proposed model applies least squares, cross-sectional, multiple regression analysis, considering both fundamental factors and technical factors. The portfolio constructor with the new expected return model is tested on the real stock market. Results show that the performance of the new model is superior to the previous one which is based on CAPM model.