An Application of Optimal Stock Selection Strategies and Capital Allocation Strategies by Genetic Algorithm combines with the Rule Base of Indicators Analysis Based on Communications and Internet

• Main Authors: Jhe-An Syu, 許哲銨
• Other Authors: Chia-Hsuan Yeh
• Format: Others
• Language: zh-TW
• Published: 2010
• Online Access: http://ndltd.ncl.edu.tw/handle/49372087389103491253

碩士 === 元智大學 === 資訊管理學系 === 98 === Stock market investment is the most popular investment tool for public. Issues in investment decision-making, three strategies are the direction of major research, including stock selection strategies, technical timing strategies, and capital allocation strategies. The relationships between the three strategies are complex, so the three strategies were studied independently to simplify the question in the past. This study proposes the integration of stock-picking strategy and capital allocation strategy as a major research objective. The study uses genetic algorithms and the financial indicators rule-base to find out the optimization of portfolio and capital allocation strategy, and use technical indicators analysis rule-base to carry out timing strategy. Finally, the study will evaluate the timing and the buy-hold strategy of return on investment (ROI) on last trading day. In addition, genetic algorithms operate in this study, the chromosome presented by integer-code in portfolio problem, and combined with the string-bits to avoid evolution issues of illegal and duplicate stock selection. In this study, we use the data of network communications industry from 1999 to 2009. The results of experiments show that genetic algorithms method can effectively search for optimal financial structure of listed companies of the investment portfolio, and to use the financial analysis indicators of the past five years, the average of capital allocation strategies and the stock market-hold strategy will get a better return on investment.