臺灣博碩士論文加值系統

外匯市場是最複雜的金融市場之一。匯率常具有高雜訊、非穩態、非線性等特徵。因此在匯率預測中，使用非線性模型預測較為適當。

本論文建制了一個長短期記憶遞迴類神經網路(long short-term memory recurrent neural network)的預測模型用以預測外匯市場，該方法預測新台幣兌美元之隔日漲跌方向準確度為53.8%。而以 LSTM-RNN (with dropout) 預測各國匯率，得預測準確度最高為韓元兌美元，準確度達55.84%，最低為澳幣兌美元，準確度達49.43%。

Foreign exchange (FX) market is one of the most complex financial systems. Foreign exchange rates typically contain high noise, non-stationarity and non-linearity. As a result, it is necessary to use non-linear models for forcasting purposes.

We build a forecasting system based on the LSTM-RNN (long short-term memory recurrent neural network) with dropout model to predict FX markets. This method predicts the direction of change in USD/TWD exchange rates for the next day with 53.8% accuracy. In addition, we use the LSTM-RNN with dropout model to predict the exchange rates of other countries. The best result is the USD/JPY exchange rate, with 55.84% accuracy, and the worst result is the USD/AUD exchange rates, with 49.43% accuracy.

誌謝 iii
摘要 v
Abstract vii
1 緒論 1
1.1 簡介 1
1.2 論文架構 3
2 匯率預測相關文獻 5
2.1 效率市場假說 5
2.2 類神經網路模型預測匯率文獻回顧 6
3 類神經網路相關文獻 9
3.1 時間遞迴類神經網絡 9
3.2 損失函數與梯度下降法 10
3.3 時序性倒傳遞法 12
3.4 梯度消失問題 13
3.5 長短期記憶類神經網路 14
3.6 正規化方法:放棄法 16
4 實驗設計 9
4.1 模型概述 19
4.2 實證資料選取 21
4.3 效能測試 21
4.4 驗證資料處理 21
5 實驗結果 23
5.1 模型預測結果與比較 23
5.2 卡方適合度檢定 24
6 結論與未來展望 27
6.1 結論 27
6.2 未來展望 27
Bibliography 29

[1] M. Adya and F. Collopy. How effective are neural networks at forecasting and prediction? a review and evaluation. Journal of Forecasting, 17(5-6):481–495, 1998.
[2] Y. Bengio, P. Simard, and P. Frasconi. Learning long-term dependencies with gradient descent is difficult. IEEE Transactions on Neural Networks, 5(2):157–166, 1994.
[3] L. Bottou and O. Bousquet. The tradeoffs of large scale learning. In Proceedings of the 20th International Conference on Neural Information Processing Systems(NIPS), pages 161–168, Vancouver, Canada, 2007.
[4] G. Cybenko. Approximation by superpositions of a sigmoidal function. Mathematics of Control, Signals, and Systems, 2(4):303–314, Dec. 1989.
[5] J. Donahue, L. A. Hendricks, M. Rohrbach, S. Venugopalan, S. Guadarrama, K. Saenko, and T. Darrell. Long-term recurrent convolutional networks for visual recognition and description. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(4):677–691, 2017.
[6] R. Engle. GARCH 101: the use of ARCH/GARCH models in applied econometrics. Journal of Economic Perspectives, 15(4):157–168, 2001.
[7] E. F. Fama. Efficient capital markets: a review of theory and empirical work. Journal of Finance, 25(2):383–417, 1970.
[8] C. L. Giles, S. Lawrence, and A. C. Tsoi. Noisy time series prediction using recurrent neural networks and grammatical inference. Machine Learning, 44(1):161–183, 2001.
[9] A. Graves. Generating sequences with recurrent neural networks. arXiv preprint arXiv:1308.0850, 2013.
[10] T. H. Hann and E. Steurer. Much ado about nothing? exchange rate forecasting: neural networks vs. linear models using monthly and weekly data. Neurocomputing, 10(4):323–339, 1996.
[11] S. Hochreiter, Y. Bengio, P. Frasconi, and J. Schmidhuber. Gradient flow in recurrent nets: the difficulty of learning long-term dependencies. In S. C. Kremer and J. F. Kolen, editors, A Field Guide to Dynamical Recurrent Neural Networks. IEEE Press, New York, 2001.
[12] S. Hochreiter and J. Schmidhuber. Long short-term memory. Neural Computation, 9(8):1735–1780, 1997.
[13] W. Huang, K. K. Lai, Y. Nakamori, and S. Wang. Forecasting foreign exchange rates with artificial neural networks: a review. International Journal of Information Technology and Decision Making, 3(1):145–165, 2004.
[14] L. Ingber. Statistical mechanics of nonlinear nonequilibrium financial markets. Mathematical and Computer Modelling, 23(7):101–121, 1996.
[15] D. P. Kingma and J. Ba. Adam: a method for stochastic optimization. In Proceedings of the 3rd International Conference on Learning Representations (ICLR), San Diego, May 2015.
[16] A. Lapedes and R. Farber. Nonlinear signal processing using neural networks: prediction and system modeling. Technical Report LA-UR-87-2662, Los Alamos National Laboratory, Los Alamos, NM, 1987.
[17] A. Lo and A. MacKinlay. A non-random walk down Wall Street. Princeton University Press, Princeton, NJ, 1999.
[18] B. Malkiel. A random walk down Wall Street. Norton, New York, 1996.
[19] P.-F. Pai and C.-S. Lin. A hybrid arima and support vector machines model in stock price forecasting. Omega, 33(6):497–505, 2005.
[20] R. Pascanu, T. Mikolov, and Y. Bengio. On the difficulty of training recurrent neural networks. In Proceedings of the 30th International Conference on International Conference on Machine Learning, pages 1310–1318, Atlanta, June 2013.
[21] N. Qian. On the momentum term in gradient descent learning algorithms. Neural Networks, 12:145–151, 1999.
[22] J. Snyman. Practical mathematical optimization: an introduction to basic optimization theory and classical and new gradient-based algorithms. Springer, Berlin, 2005.
[23] N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov. Dropout: a simple way to prevent neural networks from overfitting. Journal of Machine Learning Research, 15:1929–1958, 2014.
[24] I. Sutskever, J. Martens, and G. Hinton. Generating text with recurrent neural networks. In Proceedings of the 28th International Conference on Machine Learning(ICML-11), pages 1017–1024, Bellevue, WA, June 2011.
[25] I. Sutskever, O. Vinyals, and Q. V. Le. Sequence to sequence learning with neural networks. In Proceedings of the 27th International Conference on Neural Information Processing Systems (NIPS-14), pages 3104–3112, Montreal, December 2014.
[26] R. S. Sutton. Two problems with backpropagation and other steepest-descent learning procedures for networks. In Proceedings of the 8th Annual Conference of the Cognitive Science Society (CogSci), pages 823–831, Amherst, MA, 1986.
[27] S. Taylor. Modelling financial time series. John Wiley & Sons, Chichester, UK, 1986.
[28] F.-M. Tseng, G.-H. Tzeng, H.-C. Yu, and B. J. C. Yuan. Fuzzy arima model for forecasting the foreign exchange market. Fuzzy Sets and Systems, 118(1):9–19, 2001.
[29] O. Vinyals, A. Toshev, S. Bengio, and D. Erhan. Show and tell: a neural image caption generator. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 3156–3164, Boston, MA, June 2015.
[30] D. H. Wolpert and W. G. Macready. No free lunch theorems for optimization. Transactions on Evolutionary Computation, 1(1):67–82, 1997.
[31] J. Yao and C. L. Tan. A case study on using neural networks to perform technical forecasting of forex. Neurocomputing, 34(1-4):79–98, 2000.
[32] G. Zhang, B. Eddy Patuwo, and M. Y. Hu. Forecasting with artificial neural networks: the state of the art. International Journal of Forecasting, 14(1):35–62, 1998.