Loading...
|
Please use this identifier to cite or link to this item:
https://nccur.lib.nccu.edu.tw/handle/140.119/146603
|
| Title: | 基於神經網路模型預測的外匯交易策略
Forex Trading Strategies Based on Neural Network Model |
| Authors: | 王瑞杰
Wang, Ruijie |
| Contributors: | 張興華
Chang, Hsing-Hua
王瑞杰
Wang, Ruijie |
| Keywords: | 外匯市場
神經網路
卷積神經網路
長短期記憶模型
CNN-LSTM
Forex market
Neural network
CNN
LSTM
CNN-LSTM |
| Date: | 2023 |
| Issue Date: | 2023-08-02 14:12:01 (UTC+8) |
| Abstract: | 貨幣市場作為重要的金融市場，機器學習在該場景的應用多為基於對單獨貨幣對的預測而執行策略，鮮少有將基於機器學習預測的交易策略運用在貨幣截面上。本論文以美國投資者為視角，使用1998至2022年剔除掛鉤貨幣的23個國家貨幣作為樣本，通過CNN、LSTM、CNN-LSTM模型同時預測所有貨幣樣本走勢，並形成做多前25%投組做空後25%投組的交易策略，嘗試探討一下幾個問題：1）基於神經網路模型預測的交易策略是否在貨幣截面上產生報酬；2）相較單獨的傳統交易策略（動能交易、利差交易、動能反轉交易、價值交易）以及基於傳統交易因子的OLS模型預測是否產生更高的截面報酬；3）比較針對圖像的CNN模型和針對時序資料的LSTM模型的績效，同時比較CNN-LSTM混合模型是否結合兩者特質而做出更準確的預測。
本論文發現基於LSTM預測的截面貨幣交易策略取得最好的超額報酬，2019-2022年間的平均年化報酬率為6.62%，且優於任何單獨的傳統交易策略，其中動能交易策略在預測期間失效，原因可能是整體貨幣樣本在此期間的疲軟導致。同時三個神經網路模型都展現出較OLS模型展現出更高的績效，而CNN-LSTM混合模型並未展現出結合CNN和LSTM優勢的效果。
The foreign exchange market is one of the essential parts of the financial market, the application of machine learning in the foreign exchange market is mostly based on the prediction of individual currency pairs and executing the strategies, but few applications of trading strategies based on machine learning predictions on the cross-section of currencies. From the perspective of American investors, this paper uses 23 national currencies excluding pegged currencies from 1998 to 2022 as samples, simultaneously predicts the trends of all currency samples through CNN, LSTM, and CNN-LSTM models, and long the highest 25% portfolio and short the lowest 25% portfolio. This paper attempts to explore the following issues: 1) whether trading strategies based on neural network model predictions generate returns on the cross-section of currencies; 2) compared with individual traditional trading strategies (Momentum trade, Carry trade, Momentum reversal trade, Value trade) and predictions based on OLS models of traditional trading factors, whether they generate higher cross-sectional returns; 3) compare the performance of CNN models for images and LSTM models for time series data, and compare whether CNN-LSTM hybrid models combine the characteristics of both to make more accurate predictions.
This paper finds that the cross-sectional currency trading strategy based on LSTM predictions achieves the best excess returns, with an average annual return rate of 6.62% from 2019 to 2022, and performs better than any individual traditional trading strategy. The momentum trading strategy failed during the prediction period, which may be due to the overall weakness of the currency sample during this period. Also, all three neural network models show higher performance than the OLS model, but the CNN-LSTM hybrid model does not show the effect of combining the advantages of CNN and LSTM. |
| Reference: | Asness, C. S., Moskowitz, T. J., & Pedersen, L. H. (2013). Value and Momentum Everywhere. The Journal of Finance, 68(3), 929-985.
Andy C. W. Chui, Titman, S., & John Wei, K. C. (2010). Individualism and Momentum around the World. The Journal of Finance, 65(1), 361-392.
Burnside, C., Eichenbaum, M., & Rebelo, S. (2011). Carry Trade and Momentum in Currency Markets. Annual Review of Financial Economics, 3, 511-535
Chaboud, A. P., & Wright, J. H. (2005). Uncovered interest parity: It works, but not for long. Journal of International Economics, 66(2), 349-362.
Chen, K., Zhou, Y., & Dai, F. (2015). A LSTM-based method for stock returns prediction: A case study of China stock market. In 2015 IEEE International Conference on Big Data, Santa Clara, CA, USA, 2015, pp. 2823-2824,
Cumby, R. E., & Obstfeld, M. (1981). A Note on Exchange-Rate Expectations and Nominal Interest Differentials: A Test of the Fisher Hypothesis. The Journal of Finance, 36(3), 697-703.
Daniel, K., Hirshleifer, D., & Subrahmanyam, A. (1998). Investor Psychology and Security Market Under- and Overreactions. The Journal of Finance, 53(6), 1839-1885.
Di Persio, L., & Honchar, O. (2016). Artificial neural networks architectures for stock price prediction: Comparisons and applications. International journal of circuits, systems and signal processing, 10, 403-413.
Doskov, N., & Swinkels, L. (2015). Empirical evidence on the currency carry trade, 1900–2012. Journal of International Money and Finance, 51, 370-389.
Fama, E.F. (1984) Forward and Spot Exchange Rates. Journal of Monetary Economics, 14, 319-338.
Fama, E. F., & French, K. R. (1992). The Cross-Section of Expected Stock Returns. The Journal of Finance, 47(2), 427-465.
Galeshchuk, S., & Mukherjee, S. (2017). Deep networks for predicting direction of change in foreign exchange rates. Intelligent Systems in Accounting, Finance and Management, 24(4), 100-110.
Hansen, L. P., & Hodrick, R. J. (1980). Forward Exchange Rates as Optimal Predictors of Future Spot Rates: An Econometric Analysis. Journal of Political Economy, 88(5), 829-853
Hoseinzade, E., & Haratizadeh, S. (2019). CNNpred: CNN-based stock market prediction using a diverse set of variables. Expert Systems with Applications, 129, 273-285.
Ismailov, A., & Rossi, B. (2018). Uncertainty and deviations from uncovered interest rate parity. Journal of International Money and Finance, 88, 242-259.
Jegadeesh, N., & Titman, S. (1993). Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency. The Journal of Finance, 48(1), 65-91.
Jiang, X., Han, L., & Yin, L. (2019). Currency strategies based on momentum, carry trade and skewness. Physica A: Statistical Mechanics and its Applications, 517, 121-131.
Krizhevsky, A., Sutskever, I., Hinton, G. E. (2012). ImageNet Classification with Deep Convolutional Neural Networks
Kroencke, T. A., Schindler, F., & Schrimpf, A. (2014). International Diversification Benefits with Foreign Exchange Investment Styles. Review of Finance, 18(5), 1847-1883.
Kumar, S. (2019). Does risk premium help uncover the uncovered interest parity failure? Journal of International Financial Markets, Institutions and Money, 63, 101-135.
LeCun, Y., Boser, B., Denker, J. S., Hendersn, H., Howard, R. E., Hubbard, W., Jackel, L. D.(1989).Backpropagation Applied to Handwritten Zip Code Recognition. Neural Computation, 1(4), 541-551.
Lustig, H., Roussanov, N., & Verdelhan, A. (2011). Common Risk Factors in Currency Markets. The Review of Financial Studies, 24(11), 3731-3777.
M. De Bondt, W. F., & Thaler, R. (1985). Does the Stock Market Overreact? The Journal of Finance, 40(3), 793-805.
Meese, R. A., & Rogoff, K. (1983). Empirical exchange rate models of the seventies: Do they fit out of sample? Journal of International Economics, 14(1-2), 3-24.
Menkhoff, L., Sarno, L., Schmeling, M., & Schrimpf, A. (2012). Currency momentum strategies. Journal of Financial Economics, 106(3), 660-684.
Nelson, D. M. Q. Pereira, A. C. M. & R. A. de Oliveira. (2017). Stock market`s price movement prediction with LSTM neural networks. International Joint Conference on Neural Networks (IJCNN), Anchorage, AK, USA, 2017, pp. 1419-1426
Qi, L., Khushi, M., & Poon, J. (2020). Event-Driven LSTM For Forex Price Prediction. In 2020 IEEE Asia-Pacific Conference on Computer Science and Data Engineering, Gold Coast, Australia, 2020, pp. 1-6.
Raza, A.(2015). Are Value Strategies Profitable in the Foreign Exchange Market? In 28th Australasian Finance and Banking Conference
Rosenberg, B., Reid, K. and Lanstein, R. (1985). Persuasive Evidence of Market Inefficiency. Journal of Portfolio Management, 11, 9-17.
Yıldırım, D. C., Toroslu, I. H., & Fiore, U. (2021). Forecasting directional movement of Forex data using LSTM with technical and macroeconomic indicators. Financial Innovation, 7(1), 1-36. |
| Description: | 碩士
國立政治大學
金融學系
110352035 |
| Source URI: | http://thesis.lib.nccu.edu.tw/record/#G0110352035 |
| Data Type: | thesis |
| Appears in Collections: | [金融學系] 學位論文
|
Files in This Item:
| File |
Size | Format | |
|---|
| index.html | 0Kb | HTML2 | 93 | View/Open |
|
All items in 政大典藏 are protected by copyright, with all rights reserved.
|
