Empirical Study on Trend-Factor Trading Strategies in Taiwanese Stock Market with Markov Regime Switch Regression Models

• Main Authors: YI-TING KO, 柯宜廷
• Other Authors: Jen-Jsung Huang
• Format: Others
• Language: zh-TW
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/cu456b

碩士 === 國立中山大學 === 財務管理學系研究所 === 106 === In the financial markets, investors often use technical, fundamental and macroeconomic factors to choose profitable stocks. This paper tries to use the above three factors to construct an outperforming portfolio. First, in the technical analysis, we construct the trend factor with short-term and long-term moving averages to forecast stock returns. Second, we use the hidden Markov regime switch model to forecast market return with macroeconomic factors. Finally, we construct seven investment strategies by using trend factor, profit factor and macroeconomic factor. We also decide the forecasted results of hidden Markov regime switch model in eleven ways. The empirical results show that the best performing investment strategy is to long the portfolio with 20 stocks selected by both the trend factor and the profit factor when market return is expected to rise and to short TAIEX when market return is expected to fall. This paper found that the portfolio constructed by considering the technical, fundamental and macroeconomic factors together can outperform other six portfolios.