近年來,臺灣股市成為了主要的投資管道。而未來的消息面、基礎建設的方向與外資走向都是投資者選擇股票的消息來源。雖然建立一個高效率的股市預測決策支援系統是相當困難,但如果能透過此決策系統給予投資者額外的資訊,投資者也許能夠從股票市場中賺取更多的獲利。本研究利用線段切割法(Piecewise Linear Representation-PLR)與動態時間校正法(Dynamic Time Warping-DTW)來進行股票轉折點的預測。線段切割法主要將歷史股價作細部線段切割,以判斷未來股價走勢,然而如何決定細部線段切割之最佳門檻值為一重要關鍵;而動態時間校正法主要是將目前需要判斷的股價片段取出,再利用DTW在過去的歷史資料中尋找相似的歷史片段來當作倒傳遞類神經網路(Back-propagation Neural Network, BPN)之學習片段,透過此方式加強BPN對目前趨勢調適出最佳權重。 因此,本研究利用動態時間校正法(DTW)與線段切割法(PLR)結合倒傳遞類神經網路(BPN),建構動態時間校正結合線段切割法,形成一交易決策支援系統,協助投資人為一個明確的投資考量指標,以選擇買賣股票的適當時機,有效地降低投資風險並提高報酬。 其中,倒傳遞類神經網路主要在訓練技術指標(Input variables)與買賣點(Output variables)之連結權重,基因演算法(Genetic Algorithms, GA)用來演化較佳的切割門檻值,期望能找出未來較佳買賣點時機。本研究以台灣股市與美國股票證劵交易市場中的個股為研究對象,將以預測出的買賣點進行實際的投資獲利計算,經實驗結果證明,本研究所建構之動態時間校正結合線段切割法預測模型比傳統的技術指標方法能更加準確的預測出股價走勢之最佳買賣點。
The stock market has become the main outlet for investment recently in Taiwan. The futures indicator, investment foundations, foreign capitals are diverse choices for investors. How to establish an efficient decision support system on stock market prediction is pretty difficult but it’s necessary. By adopting a good forecasting system, investors may earn more money form the stock market. This research suppose a piecewise linear representation method with a Dynamic Time Warping window for stock turning points detection. The piecewise linear representation method is able to generate numerous local stock turning points from the historic data, then the Dynamic Time Warping window will be applied to retrieve similar stock price patterns from historic data for training the system. These turning points represent short-term trading signals for selling or buying stocks from the market. A Back-Propagation neural network (B.P.N) is further applied to learn the connection weights from these historic turning points and afterwards it is applied to forecast the future turning points from the set of test data. Experimental results demonstrate that the system integrating PLR and neural networks can make a significant amount of profit when compared with other approaches using stock data such as Taiwan Stock Exchange Market, Google, Lockheed Martin Corporation, Delta Air Lines , and Caterpillar Inc., etc. , to be research object and compute actual return on investment by utilizing the prediction model, DTW-PLR. This research result can make a significant amount of profit when compared with other approaches using stock data to predict better commerce occasions (trading points) than experts’ judgment from technical indexes or other prediction models by experimental results.