外匯市場是全球金融體系當中重要的一環，利差交易為投機客、外匯交易員、避險基金等在外匯市場當中普遍採用的交易策略。近年來，新興市場貨幣在利差交易中所佔份額逐漸增加，由於新興市場利率普遍高於成熟市場，其獲利可能性使得外匯市場參與者蜂擁而至，然而，匯率之劇烈波動卻可能侵蝕掉賺取的利潤。
傳統利率模型僅以遠期溢價預測匯率變動，但在實證結果上卻往往與理論相悖離，因此，本研究將國家風險因素納入考量，並將機器學習中的類神經網路概念及技術引入金融領域當中，以長短期記憶模型(LSTM)對未來匯率變動進行預測，同時也將其預測能力與傳統迴歸模型之預測效果相互比較。實證發現，在新興市場國家當中使用LSTM神經網路模型，並以考量國家風險因子之遠期溢價預測未來匯率走勢有較傑出的預測效果。除了將類神經網路用以預測匯率變動之外，我們一樣能將人工智慧的技術應用於其他金融商品之價格預測上，隨著技術日漸進步、人工智慧相關領域的研究逐年倍增，數以萬計的應用場景將在人工智慧的環境下得以升級。

The foreign exchange market plays an important role in the global financial system. Carry Trade is one of the most popular trading strategies for speculators, forex traders, hedge funds, etc. In recent years, emerging market currencies gained market share gradually. Owing to the interest spread between emerging markets and developed markets, lots of foreign exchange market participants attracted by the profitability. However, the volatility of exchange rates might cause profit erosion.
Uncovered Interest Rate Parity (UIP) only use the forward premium to predict the changes in the foreign exchange rate, but empirical results often deviate from the theoretical studies. Therefore, this study introduces the country risk factor into the UIP equation. Adopting the concept of neural networks into the financial field, we use Long Short-Term Memory (LSTM) neural network for the foreign exchange rate prediction. Then, compares its predictive ability to the traditional regression model’s. According to the empirical study, we predict the future trend of the foreign exchange rate in the emerging markets. By using the forward premium with the country risk, LSTM neural network shows the outstanding result. Besides the implementation of currency forecasting via neural networks, Artificial Intelligence (AI) technologies can also apply to other financial products’ price prediction. With the advances of technology and AI, thousands of application scenarios will be able to be promoted.

1. Anker, P. (1999) “Uncovered Interest Parity, Monetary Policy and Time-Varying Risk Premia” Journal of International Money and Finance, Elsevier, Vol. 18(6), Pages 835-851.
2. Bengoechea, A.G., Uretaz, C. O., Saavedra, M. M., & Medina, N. O. (1996) “Stock Market Indices in Santiago de Chile: Forecasting Using Neural Networks” IEEE International Conference on Neural Networks , Vol. 4, Pages 2172-2175
3. Bildirici, M., & Ersin, Ö. Ö. (2009) “Improving Forecasts of GARCH Family Models with the Artificial Neural Networks” Journal Expert Systems with Applications: An International Journal archive, Vol. 36, Issue 4
4. Burnside, C., Eichenbaum, M., & Rebelo, S. (2008) “Carry Trade: The Gains of Diversification” Journal of the European Economic Association, Vol. 6, Issue 2-3, Pages 581–588
5. Cavallo, M. (2006) ”Interest Rates, Carry Trades, and Exchange Rate Movements” FRBSF Economic Letter
6. Chinn, M. D., & Meredith, G. (2004) “Monetary Policy and Long-Horizon Uncovered Interest Parity” IMF Staff Papers – Vol. 51, No 3
7. Fama, E. F. (1984) “Forward and Spot Exchange Rates” Journal of Monetary Economics, Vol. 14, Issue 3, Pages 319-338
8. Fischer, T., & Krauss, C. (2017) “Deep Learning with Long Short-Term Memory Networks for Financial Market Predictions” European Journal of Operational Research, Vol. 270, Issue 2, Pages 654-669
9. Froot, K. A., & Thaler, R. H. (1990) “Anomalies: Foreign Exchange.” Journal of Economic Perspectives, 4(3):179-192.
10. Hochreiter, S., & Schmidhuber, J. (1997) “Long Short-Term Memory” Neural Computation 9(8): 1735-80
11. Hutchison, M., & Sushko, V. (2013) “Impact of Macro-Economic Surprises on Carry Trade Activity” Journal of Banking & Finance, Vol. 37, Issue 4, Pages 1133-1147
12. Jylhä, P., Lyytinen, J. P., & Suominen, M. (2008) “Arbitrage Capital and Currency Carry Trade Returns” AFA 2009 San Francisco Meetings Paper
13. Kimoto, T., & Asakawa, K. (1990) “Stock Market Prediction System with Modular Neural Networks” 1990 IJCNN International Joint Conference on Neural Networks, Vol. 1, Pages 1-6
14. Lustig, H., Roussanov, N., & Verdelhan, A. (2011) “Common Risk Factors in Currency Markets” The Review of Financial Studies, Vol. 24 ,Issue 11, Pages 3731-3777
15. Lustig, H., & Verdelhan, A. (2005) “The Cross-Section of Currency Risk Premia and US Consumption Growth Risk” NBER Working Paper No. 11104
16. MacDonald, R., & Torrance, T. S. (1989) “Some Survey Based Tests of Uncovered Interest Parity” In MacDonald, R. and Taylor, M.P. (eds.) Exchange Rates and Open Economy Macroeconomics, Pages 239-248
17. Namin, S. S., & Namin, A. S. (2018) “Forecasting Economic and Financial Time Series: ARIMA VS. LSTM”
18. Persio, L. D., & Honchar, O. (2017) “Recurrent Neural Networks Approach to the Financial Forecast of Google Assets” International Journal of Mathematics and Computers in Simulation, Vol. 11, Pages 7-13
19. Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1985) “Learning Internal Representations by Error Propagation” DTIC Document, Tech. Rep.
20. Santos, M. B. C., Klotzle, M. C., & Pinto, A. C. F. (2016) “Evidence of Risk Premiums in Emerging Market Carry Trade Currencies” Journal of International Financial Markets, Institutions & Money, Vol. 44, Pages103-115
21. Tseng, F. M., Yu, H. C., & Tzeng, G. H. (2002) “Combining Neural Network Model with Seasonal Time Series ARIMA Model” Technological Forecasting & Social Change, Vol. 69, Issue 1, Pages 71–87
22. Yim, J. (2002) “A Comparison of Neural Networks with Time Series Models for Forecasting Returns on a Stock Market Index.” Lecture Notes in Computer Science, Pages 25-35