Stock Trading Support Using Dynamic Reduct in Rough Sets

• Main Authors: Yeh-Ching, 葉青
• Other Authors: 王派洲
• Format: Others
• Language: zh-TW
• Published: 2010
• Online Access: http://ndltd.ncl.edu.tw/handle/96579276372066878414

碩士 === 南台科技大學 === 國際企業系 === 98 === ABSTRACT In recently years, people pay more and more attention in investment. When people holding extra income, they often consider to investment funds ,deposit, stocks, bonds, rotating Savings and credit Associations, real estate, gold, foreign exchange, futures or other financial investment methods to earn reward or capital appreciation. However, stock market is an open and dynamic environment but it is easily influenced by political, economical, human, or other factors. If investors can catch the trend of the selected target, they will have larger chance to gain more profit. Our data is collected from July 1, 2002, to December 29, 2009, (The search period is from July 1, 2002, to December 31, 2008, where the testing periods is from January 5, 2009, to December 31, 2009). In this research, we selected twelve stocks as our research subjects and 10 technical indicators with TAIEX are used as technical indicators. In this research, we combine genetic algorithm and rough set theory as our predicting engine and we look forward to build a system to assist investors in trading. We wish we can accurately predict the trading points and make the right decision on these trading points. The empirical results show that the twelve stocks are still surpass 74% above yield rate during the one year testing period from January 5, 2009, to December 31, 2009. The highest yielding even reaches as high as 176%. Therefore, accession to TAIEX and the principle of the application will help forecast share trading effectively.