碩士 / 國立中正大學 / 財務金融系研究所 / 107 / With the previous studies, head and shoulders has been proven is a profitable strategy, but only have around 60% of the profitability in one year. This paper use three method to raise head and shoulders strategy, Logistic Regression, Random Forest, Artificial Neural Networks with 24 factors chosen from previous researches. The result shows that Artificial Neural Networks is best method upon three methods with highest accuracy in average and precision, recall. The factors that affect the success or failed signals is about the risks and returns. This paper also combines the use empirical data to buy and sell. The result shows ANN can choose the most profitable stock from the head and shoulders pattern.
| Identifer | oai:union.ndltd.org:TW/107CCU00304041 |
| Date | January 2019 |
| Creators | CHAN, SHAN-KAI, 詹昇凱 |
| Contributors | CHEN, AN-SING, 陳安行 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | en_US |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 57 |
Page generated in 0.0175 seconds