基因參數最佳化技術指標之集成策略研究-以股市大盤為例

現時一般大眾皆已了解投資理財的重要性，在眾多理財工具中又以股票市場之進入門檻較低、資訊易取得、發展悠久建全，而成為最熱門的投資標的，本篇研究有鑑於一般大眾難以精準掌握進出場時機，追高、套牢、認賠出場等屢見不鮮，因此將對技術分析擇時的部份進行探討，期望找出能準確判斷適當買賣時機的一套方法，並能夠廣泛使用。本研究應用基因演算法（Genetic Algorithms, GA）對現行技術指標所用之天數、買賣門檻等參數進行最佳化，改善一般投資者常用的傳統技術指標，使其能更符合市場環境。接著再將最佳化參數後的各種技術指標，利用集成（Ensemble）的方式進行共同決策，以提升策略穩定度及準確度。最後將各種策略套用至多國股市，進行虛擬交易進行實證。實驗結果顯示，技術分析於各國市場皆為有效，單一技術指標透過集成後，可有效提升穩定度，並超越買入持有策略，成為良好的投資策略。

關鍵字

基因演算法；技術指標；集成；股市擇時

並列摘要

Generally, everyone has been realized the importance of financial investment. The reasons why stock investment is getting more popular are its low threshold, mass information, and developed system. The techniques in stock market are divided into two categories, selectivity and timing; that is, how to choose stocks (basic analysis) and when sharing and selling them accurately (technical analysis). It is difficult for most stockholders to judge proper timing for sharing and selling that they usually lose tremendous amount of money because of trap or some wrong decisions. For this reason, depending on technical analysis, this research focuses on finding methods that stockholders could decide precisely and use them extensively in stock market. First, Genetic Algorithms (GA) is conducted in this study which is ideal for current market if the index could improve data such as numbers of days or trading threshold and traditional index used in the past. Next, to enhance the stability and accuracy of strategies, the improved data coming from the index are processed in common decision in terms of Ensemble. Finally, each strategy is examined based on virtual transaction in various foreign stock markets.

並列關鍵字

Genetic Algorithms ； Technical Index ； Ensemble ； Stock Market Timing

參考文獻

[2] F. Allen, R. Karjalainen, “Using genetic algorithms to find technical trading rules,” Journal of Financial Economics, vol. 51, 1999, 245-271.

[3] A. Gupta, and A. Sivakumar, “Simulation based multiobjective schedule optimization in semiconductor manufacturing,” Proceeding of the 2002 Winter Simulation Conference, vol. 2, 2002, pp.1862-1870.

[4] Brock, William and Lakonishok, Josef & LeBaron, Blake, "Simple Technical Trading Rules and the Stochastic Properties of Stock Returns," Journal of Finance, American Finance Association, vol. 47, no. 5, 1992, pp. 1731-64

[7] G. Garai, and B. B. Chaudhuri, “A novel genetic algorithm for automatic clustering,” Pattern Recognition Letters, vol. 25, 2004, pp. 173-187.

[9] A. Gunasekarage, and D.M. Power, “The Profitability of Moving Average Trading Rules in South Asian Stock Markets,” Emerging Markets Review, vol. 2, no. 1, 2001, 17-33.

國際替代計量

基因參數最佳化技術指標之集成策略研究-以股市大盤為例

全文下載

主題瀏覽