應用獨立成份分析與支援向量迴歸於財務時間序列預測

Using Independent Component Analysis and Support Vector Regression for Financial Time Series Forecasting

呂奇傑(Chi-Jie Lu) ； 李天行(Tian-Shyug Lee) ； 陳學群(Hsueh-Chun Chen)

《資訊管理學報》 14卷4期 (2007/10) Pp. 161-183

https://doi.org/10.6382/JIM.200710.0007

摘要

由於財務時間序列資料具有高頻率、雜訊、非定態與混沌等性質，因此在現今時間序列預測領域中，向來被認為是一極具挑戰性的應用領域。本研究提出一結合獨立成份分析(Independent component analysis, ICA)與支援向量迴歸(Support Vector Regression, SVR)之財務時間序列預測模式，先利用ICA具有將混合訊號分離出個別獨立之來源訊號之能力，從預測變數中估計出獨立成份，並在去除代表雜訊的獨立成份及保留主要的獨立成份作為預測變數後，再使用SVR以濾除雜訊後之預測變數建構預測模式。期望可以讓SVR在建構模式時不受雜訊影響，進而提升預測結果的準確度。為驗證所提方法之有效性，本研究將以日經225現貨開盤指數及台股現貨收盤指數之預測進行實證研究，並與單純使用SVR模式及隨機漫步模式的預測結果作比較。實證結果顯示，所提之方法不論是在預測誤差或是預測準確度的表現上均較單純使用SVR及隨機漫步模式為佳。

關鍵字

獨立成份分析 ； 支援向量迴歸 ； 財務時間序列預測 ； 股價指數

並列摘要

As financial time series are inherently noisy, non-stationary and deterministically chaotic, it is one of the most challenging applications of modern time series forecasting. Due to the advantages of the generalization capability in obtaining the unique and global optimal solution, support vector regression (SVR), has also been successfully applied in time series prediction, especially in the financial time series forecasting. In the modeling of financial time series using SVR, one of the key problems is the inherent high noise. Therefore, detecting and removing the noise are important but difficult tasks when building an SVR forecasting model. To alleviate the influence of noise, a two-stage approach by integrating independent component analysis (ICA) and support vector regression is proposed in this research for financial time series forecasting. ICA is a novel statistical signal processing technique that was originally proposed to find the latent source signals from observed mixture signal without knowing any prior knowledge of the mixing mechanism. The proposed approach first uses ICA to the forecasting variables for generating the independent components (ICs). After identifying and removing the ICs containing the noise, the rest of the ICs are then used to reconstruct the forecasting variables which contain less noise. The SVR is then applied to use the filtered (or denoised) forecasting variables to build the forecasting model. In order to evaluate the performance of the proposed approach, the Nikkei 225 opening index is used as the illustrative example. The experimental results show that the proposed model outperforms the SVR model with non-filtered forecasting variables and random walk model.

並列關鍵字

Independent component analysis ； Support vector regression ； Financial time series forecasting ； Stock index

參考文獻

Antoniou, A.,Holmes, P.(1995).Futures Trading, Information and Spot Price Volatility: Evidence for the FTSE-100 Stock Index Futures Contract Using GARCH.Journal of Banking and Finance.19,117-129.

Back, A.,Weigend, A.(1997).Discovering Structure in Finance Using Independent Component Analysis.Proceeding of 5th International Conference on Neural Networks in Capital Market.(Proceeding of 5th International Conference on Neural Networks in Capital Market).

Balachandher, K.G.,Fauzias, M.N.,Lai, M.M.(2002).An Examination of the Random Walk Model and Technical Trading Rules in the Malaysian Stock Market.Quarterly Journal of Business & Economics.41,81-104.

Barlett, M.,Sejnowski, T.J.(1997).Viewpoint Invariant Face Recognition Using Independent Component Analysis and Attractor Networks.Advances in Neural Information Processing Systems.9,817-823.

Bartlett, M.S.,Movellan, J.R.,Sejnowski, T.J.(2002).Face Recognition By Independent Component Analysis.IEEE Transactions on Neural Networks.13,1450-1464.