捲積類神經網路與遞歸類神經網路預測類股趨勢能力之比較

A Comparison of Convolution Neural Network and Recurrent Neural Network on Predicting Sector Trends

陳啟元(Chi-Yuan Chen)

指導教授 ： 呂育道

國立臺灣大學/電機資訊學院/資訊工程學研究所/碩士(2019年)

https://doi.org/10.6342/NTU202000002

摘要

深度學習於近年來在各領域皆有廣泛的應用，而捲積類神經網路和長短期記憶類神經網路為其中兩大分支。
通常大家選擇使用長短期記憶類神經網路處理序列建模任務，如金融時間序列預測、語音辨識、音樂生成等。然而，一些研究逐漸開始使用捲積類神經網路處理此類任務並獲得不錯的效果。我們比較兩種類神經網路在台灣股票市場電子類股指數漲跌預測的準確率，可以看出在以日為單位的預測上，長短期記憶類神經網路表現較佳。

關鍵字

捲積類神經網路 ； 長短期記憶神經網路 ； 台灣證券市場 ； 股價預測

並列摘要

Recently deep learning has been widely applied in diverse fields. The convolutional neural network (CNN) and the long short-term memory neural network (LSTM) are two main branches in deep learning.
Previously, LSTM was the obvious candidate in tackling sequence modeling tasks, such as financial time series prediction, speech recognition, music generation, and so on. But the CNN is catching up and getting its share of attention as it also provides satisfactory results; therefore, researchers are being attracted to it. Our goal is to compare CNN and LSTM on the prediction accuracy of Taiwan Electronics Total Return Index. LSTM is found to outperform CNN with daily data.

並列關鍵字

Convolutional neural network ； Long short-term memory ； Taiwan stock market ； Stock price prediction.

參考文獻

Adebiyi, A. A., Adewumi, A. O., and Ayo, C. K. (2014). Comparison of arima and artificial neural networks models for stock price prediction. Journal of Applied Mathematics, 2014. DOI:10.1155/2014/614342

Bengio, Y., Courville, A., and Vincent, P. (2013). Representation Learning: A Review and New Perspectives. IEEE Transactions on Pattern Analysis and Machine Intelligence, 35(8),1798–1828. DOI: 10.1109/TPAMI.2013.50.

Fama, E. F. (1970). Efficient capital markets: A review of theory and empirical work. Journal of Finance, 25(2), 383–417. DOI:10.2307/2325486

He, K., Zhang, X., Ren, S., and Sun, J. (2015). Deep residual learning for image recognition. In 2016 IEEE Conference on Computer Vision and Pattern Recognition, Volume 1, Las Vegas, Nevada, USA, June 27–30, 2016 (pp. 770–778). Washington, DC: IEEE Computer Society. DOI:10.1109/CVPR.2016.90

Hochreiter, S. and Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8), 1735–1780. DOI:10.1162/neco.1997.9.8.1735