應用市場輪廓於台指期市場行為發現之研究

Abstract

本研究以市場輪廓理論為基礎，嘗試設立一指標數值作為判斷價格、價值變化與趨勢間關係之依據，同時搭配技術指標，作為倒傳遞類神經網路輸入變數，以期透過物理力量的總合評判，從中萃取市場邏輯與市場結構變化之知識規則。 為使神經網路有效學習市場輪廓知識，本研究提出定性及定量兩種市場輪廓指標計算方法，以將市場輪廓表達及轉換為輸入變數，同時為探討股價長線保護短線，短線支持長線之效益，除以長線(75分鐘前市場變化)作為市場輪廓指標計算依據外，更加入短線(15分鐘前市場變化)因素，最後透過對預測模型的績效評估，比較不同市場輪廓計算方式，與單純使用技術指標之模型間的差異性。 實驗結果顯示，以定性與定量計算方法進行比較，定性方法於預測時間較短時其績效優於定量，而定量則由於趨勢判讀能力較佳而在長時間預測上較具優勢；無論使用何種計算方式，透過加入短線市場輪廓指標有助於提升預測績效，尤以預測時間較短時之提升效果最顯著，最後由研究結果證實：相較於單純使用技術指標作為神經網路輸入變數，加入市場輪廓指標確實能有效提升預測準確度與獲利能力，因此驗證了市場輪廓的可用性。

This research applies market profile to establish an indicator to classify the correlation between the variation in price and value with the stock trends. The indicator and technical index are parameters to neural network architecture and assist to extrapolate the market logic and knowledge rules that influence the market structure via an integral assessment of physical quantities. To implement the theory of market profile on neural network architecture, this study proposes qualitative and quantitative methods to compute a market profile indicator. In addition, the indicator considers the variation and relevance between long-term and short-term trends by incorporating the long-term and short-term change in market in its calculation. A assessment of forecasting performance on different calculation approaches of market profile indicator and technical analysis is conducted to differentiate their accuracies and profitability. Experimental results show qualitative market profile indicator outperforms quantitative approach in short-term forecast period. In contrast, quantitative market profile indicator has a better trend-predicting ability thus it is more effective in long-term forecast period. The results also manifest that both approaches considering the combination of long-term and short-term change in market enhance forecasting performance and are most effective in short-term time interval. In conclusion, the integration of market profile and technical analysis surpasses technical analysis as a parameter to neural network architecture by effectively improving forecasting performance and profitability.