The Profitability of Price and Quantity Technical Trading Rules – A Comparison of Linear and Nonlinear Models

• Main Authors: Yi-Hui Lo, 羅一翬
• Other Authors: Chin-Shien Lin
• Format: Others
• Language: zh-TW
• Published: 2002
• Online Access: http://ndltd.ncl.edu.tw/handle/79615500017131335180

碩士 === 靜宜大學 === 企業管理研究所 === 90 === The market efficiency hypothesis has been an important issue in both the academic and practical fields for years. Usually people use technical analysis to test the weak-form market efficiency hypothesis. Technical analysis assumes that the history of stock price movement will repeat itself, using past stock prices and quantities to forecast the future stock price. Based on the assumptions that the technical indicators are useful, we try to build linear models, regression and GARCH, and nonlinear models, Neural Network and Neuro-Fuzzy, to investigate the predictability of the technical indexes. The empirical results show that Neuro-Fuzzy can outperform the other models no matter in bull market or bear market. Besides, Neuro-Fuzzy can also beat the market under different transaction costs. This research somehow shows the profitability of the nonlinear model of the technical indexes, which has not been explored much yet.