基於區間資訊與支援向量迴歸建構區間值股價預測模式

預測是人們想要瞭解未來的不確定事件。人們之所以做預測，是恐懼未來的不確定性與無法控制性。在以往的預測中，點預測為最常使用的預測方法，然而在生活中有許多事物存在著不明確的特性，而點預測卻無法呈現出預測結果的變異，其建構的模式不足以描述每日或每月的發展趨勢，因此區間值預測開始被眾多學者重視及廣泛應用在各個領域之中。本研究將嘗試建構出適合於股價指數的區間值預測模式，期望能提供投資人更好的決策依據。本研究將使用三種區間組成方法，分別為：上下界限法（Upper and Lower-bound method, UL）、中心半徑法（Center and Radius method, CR）及MSL法（Mean，Standard deviation and Level），使用支援向量迴歸（SVR）為預測技術，來建構區間值股價預測模式，實證對象為臺灣證券交易所（Taiwan Stock Exchange, TWSE）所編製發行量加權股價指數（TAIEX），來驗證區間資訊建構區間值股價預測模式之準確性。實證結果分析發現，預測模式最佳為加權指數資料為5秒鐘區間，並使用CR的區間組成方法。顯示對本研究所使用的股票資料而言，股價資料為5秒鐘區間，並使用中心半徑法，較為合適的預測模式。

關鍵字

股價預測；區間值預測；支援向量迴歸

並列摘要

Stock index forecasting is regarded as a challenging task of the prediction problem since the stock market is a complex, evolutionary, and nonlinear dynamic system. Interval-valued time series forecasting indicates possible future outcomes for upper and lower bounds of interval-valued data and generates a prediction interval. It has the advantage of taking into account the variability and uncertainty to reduce the amount of random variation relative to that found in classic point forecasting/ single-valued forecasting model. In this study, three interval-valued stock index forecasting models based on time-interval information and support vector regression (SVR) have been proposed. In the forecasting models, the upper and lower bounds (UL), center and radius (CR) and MSL scheme (mean, standard deviation and level) are respectively used to describe the bounds of interval-valued data and SVRs are integrated to construct stock index forecasting models. A TAIEX stock index data are used as illustrative examples to evaluate the performance of the proposed methods. Experimental results showed that the CR-based interval-valued stock index model outperforms the two competing models and hence is an effective alternative for interval-valued forecasting of stock index.