Building Industry Index Capital Allocation Model Using Genetic Algorithm

• Main Authors: Ying-Ying Lai, 賴盈穎
• Other Authors: Ping-Chen Lin
• Format: Others
• Language: zh-TW
• Published: 2010
• Online Access: http://ndltd.ncl.edu.tw/handle/22295158057075845346

碩士 === 國立高雄應用科技大學 === 金融資訊研究所 === 98 === Portfolio theory is eternal discussion of financial issues over the years, in portfolio theory, mainly method to spread investment risk and increase return on investment. How to measure the return on the investment portfolio and risk, and how the weight distribution of the portfolio has been valued by the scope for scholars. The traditional method for finding the efficient frontier is mostly linear model, but this study is the use of genetic algorithms (Genetic Algorithm, GA) solution for nonlinear evolution mechanism, and set the encoding for the real operation, no need for another code and decoding process. Besides establishing models of the many scholars also You are committed to investing measure, such Sharpe (1969) to propose Sharpe Ratio, can simple calculation, investment portfolio performance, but traditional risk need hypothesis normality for the standard deviation as a measure risk indicators used in this study is the value of portfolio risk (Portfolio Value at Risk, PVaR) approach can more accurately capture the portfolio risk to replace the standard deviation below. Use of financial insurance, electrical machinery and steel industry stock returns and the three types of index returns substitution genetic algorithm, solving the optimal portfolio weights, and the establishment of a modified Sharpe Ratio－like-Sharpe, to as genetic algorithms fitness function, in pursuit of maximizing the portfolio model－GA-PVaR. The results show that GA-PVaR model to calculate the portfolio weights, get more than the GA-MV paste into the index value.