以人工智慧為基礎股市低風險套利分析改善之研究

Abstract

現代的金融投資市場是由投機、避險與套利三種不同的系統所構成。不同的金融模型中對於市場多存有不同的假設，然而在實際的金融市場中，對於如：未來之利率、市場本身的變動無不是時時刻刻在變化，與傳統的金融理論存有矛盾之處。針對套利交易而言，市場存有：稅賦及交易成本，借貸利率差異、賣空的限制等等與理論有所出入的地方，再者對於市場的未來變化金融理論並未提出有效的方式來描述，在缺乏理論根據之下，在實際的市場之中，仍有許多的套利者的存在。但如何提供套利者以更有效率的方式在套利活動中得到合適之套利時點預測，使得套利者在套利時機出現之際以低或無風險的交易方式，獲取更大的利潤將是最重要的課題。 本研究採用電腦人工智慧領域中的類神經網路，結合不同技術分析針對股價指數進行即時套利行為的分析與趨勢預測，希望能對台灣加權股價指數的即時趨勢作一有效的預測並提供套利者更為準確的套利時點。由本研究的實証研究發現，類神經網路可以藉由學習的過程得到對於未來的趨勢變化的輸出，可以得到較佳的套利時點，獲取更多的利潤。

The contemporary market consists of the three different kinds of trading strategies, including speculation, hedge and arbitrage. Although there are many various types of finance models to describe the market, but most of them do have lots hypotheses and limitations. For example, the interest rate, the rapid change of the market itself in the real market all cause of conflicting to the finance models. Further more, there still does not exist any efficient finance theories to forecast the future market till now. There still exist arbitrageurs in the current market under these constrains. Thus, how to provide a very efficient arbitrage strategy to the arbitrageur becomes a more and more critical issue. This study applies the neural networks techniques to forecast the stock basis trend. Meanwhile, the output of the trend forecast in this study is used as a guideline to improve the arbitrage strategy. According to the result of the empirical study, the proposed neural networks enhance the lock in/out time of the arbitrage strategy.