With the deepening of economic globalization and the expanding scale of the foreign exchange market, the study of the volatility characteristics and forecasting of foreign exchange has received widespread attention.In this paper, a new forex portfolio forecasting model is proposed, which uses fully integrated empirical modal adaptive noise decomposition (CEEMDAN) to decompose the original forex price into sub-layers of different frequencies; then, convolutional neural network, bi-directional long and short term memory and attention mechanism module (CNN-BiLSTM-Attention) are used to combine the forecasting of each sub-layer; finally, the forecasting results of different frequency sub-layers are summed Reconstructed as the final prediction value. The forex series of USD, EUR, JPY and HKD against CNY were selected for empirical analysis, and based on the results of both evaluation metrics and DM tests, it was concluded that the CEEMDAN-CNN-BiLSTM-Attention model performed better in the forex market and had higher accuracy in exchange rate forecasting compared to the single model and other combined models.

Xin Li, Guoqiang Tang, Yumei Ren, Xuchang Chen. Combined CEEMDAN-CNN-BiLSTM-ATT Model for Forex Forecasting. Academic Journal of Computing & Information Science (2023), Vol. 6, Issue 7: 88-100. https://doi.org/10.25236/AJCIS.2023.060713.

[1] Minakhi Rout, Babita Majhi, Ritanjali Majhi, Ganapati Panda, Forecasting ofcurrency excha-nge rates using an adaptive ARMA model with differential evolution based training, Journal of King Saud University-Computer and Information Sciences, Volume 26, Issue1, 2014, 7-18. http://go. microsoft. com/ fwlink/p/?LinkId=255141

[2] Yu You, Xiaochun Liu. Forecasting short-run exchange rate volatility with monetary fundame-ntals: A GARCH-MIDAS approachJ, Bank. Financ. (2020). https://doi. org/10. 1016/j. jbankfin. 2020. 105849. 

[3] Stavros D, Artemis P. Multiple-Days-Ahead Value-at-Risk and Expected Shortfall Forecasting for Stock Indices, Commodities and Exchange Rates:Inter-Day versus Intra-Day Data[J], International Review of Financial Analysis, Volume 49, 2017:176-190. https://doi. org/10. 1016/j. irfa. 2016. 10. 008. 

[4] Kevin B, Aaron D. Exchange Rate Shocks and Trade:A Multivariate GARCH-M Approach[J], Journal of International Money and Finance, Volume 37, 2013: 282-305, https://doi. org/10. 1016/ j. jimonfin. 2013. 05. 010. 

[5] Katsuki Ito, Hitoshi Iima, Yoshihiro Kitamura, LSTM forecasting foreign exchange rates usin-g limit order book, Finance Research Letters, Volume 47, Part A, 2022. https://doi. org/10. 1016/j. frl. 2021. 102517. 

[6] Zhen Fang, Xu Ma, Huifeng Pan, Guangbing Yang, Gonzalo R. Arce, Movement forecasting of financial time series based on adaptive LSTM-BN network, Expert Systems with Applications, Volume 213, Part C, 2023, 119207, https://doi. org/10. 1016/j. eswa. 2022. 119207. 

[7] Pradeepta Kumar Sarangi, Muskaan Chawla, Pinaki Ghosh, Sunny Singh, P. K. Singh, FOREX tre-nd analysis using machine learning techniques: INR vs USD currenyexchange rate using ANN-GA hybrid approach, Materials Today: Proceedings, Volume 49, Part 8, 2022, Pages 3170-3176. https://doi. org/ 10. 1016/j. matpr. 2020. 10. 960. 

[8] Amit R, Avinash J. Using Bidirectional GRU and LSTM Neural Network Methods for Multi-C-urrency Exchange Rates Prediction[J]. International Journal of Innovative Technology and Exploring Engineering, 2019, 8(7). https://kns. cnki. net/kcms/detail/detail. aspx?FileName= SQFH1140BBCCB F6E3CA99B750A5BEA2511A9&DbName=GARJ2021_1

[9] Wang J Y, Wang X X, Li J Z, et al. A Prediction Model of CNN-TLSTM for USD/CNY Excha-nge Rate Prediction[J]. 2021, 9. 

[10] Abedin MZ, Moon M H, Hassan M K, et al. Deep Learning-Based Exchange Rate Prediction During the COVID-19 Pandemic. [J]. Annals of Operations Research, 2021. https://doi. org/10. 1007/S10479-021-04420-6. 

[11] Chiun-Sin Lin, Sheng-Hsiung Chiu, Tzu-Yu Lin, Empirical mode decomposition–based least squares support vector regression for foreign exchange rate forecasting, Economic Modelling, Volume 29, Issue 6, 2012, Pages 2583-2590. https://doi. org/10. 1016/j. econmod. 2012. 07. 018. 

[12] Wei Guo, Qingfu Liu, Zhidan Luo, Yiuman Tse, Forecasts for international financial series with VMD algorithms, Journal of Asian Economics, Volume 80, 2022, 101458. https://doi. org/10. 1016 /j. asieco. 2022. 101458. 

[13] Shaolong Sun, Shouyang Wang, Yunjie Wei, A new multiscale decomposition ensemble appr-oach for forecasting exchange rates, Economic Modelling, Volume 81, 2019, Pages 49-58. https://doi. org/ 10. 1016/j. econmod. 2018. 12. 013. 

[14] Yang H L, Lin H Z. Applying EMD-Based Neural Network to Forecast NTD/USD Exchange Rate[C]. The 7th International Conference on Networked Computing and Advanced Information Management, 2011, 352-357. https://ieeexplore. ieee. org/document/5967574/ citations? tabFilter =papers#citations

[15] Bhusana P, Chris T. Improving Prediction of Exchange Rates Using Differential EMD[J]. Ex-pert Systems with Applications, 2013, P377-384, https://doi. org/10. 1016/j. eswa. 2012. 07. 048. 

[16] Nur I A R, Ani S, Ruhaidah S. New Hybridization of Empirical Mode Decomposition and Le-ast Squares Support Vector Machine Model in Forecasting Malaysia Exchange Rates[J]. 2017 International Conference on Robotics, Automation and Sciences(ICORAS), 2017, pp. 1-5. https://doi. org/10. 1109/ ICORAS. 2017. 8308069

[17] Maria E T, Marcelo A C, Gaston S, et al. A Complete Ensemble Empirical Mode Decomposit-ion with Adaptive Noise[C]. Speech and Signal Processing. 2011:4144-4147. https://doi. org/10. 1109/ ICASSP. 2011. 5947265. 

[18] Lecun Y, Bottou L, Bengio Y, et al. Gradient-Based Learning Applied to Document Recognitio-n[J]. Proceedings of the IEEE, 1998, 86(11). https://doi. org/10. 1109/5. 726791. 

[19] Li W J, Qi F, Tang M, et al. Bidirectional LSTM with Self-Attention Mechanism and Multi-Channel Features for Sentiment Classification[J]. Neurocomputing, 2020, 387. https://doi. org/ 10. 1016/j. neucom. 2020. 01. 006. 

[20] Vaswani A, Shazeer N, Parmar N, et al. Attention Is All You Need[C]. arXiv, 2017. 

[21] Diebold F X, Mariano R S. Comparing Predictive Accuracy[J]. Journal of Business & Econo-mic Statistics, 2002, 20(1). 

[22] Liang Y H, Lin Y, Lu Q. Forecasting Gold Price Using A Novel Hybrid Model with ICEEM-DAN and LSTM-CNN-CBAM[J]. Expert Systems With Applications, 2022, 206. https://doi. org/10. 1016/ j. eswa. 2022. 117847. 

[23] Hum Nath Bhandari, Binod Rimal, Nawa Raj Pokhrel, Ramchandra Rimal, Keshab R. Da-hal, RajendraK. C. Khatri, Predicting stock market index using LSTM, Machine Learning with Applications, Volume 9, 2022, 100320, ISSN 2666-8270, https://doi. org/10. 1016/j. mlwa. 2022. 100320. 

[24] Umang Gupta, Vandana Bhattacharjee, Partha Sarathi Bishnu, StockNet—GRU based stock index prediction, Expert Systems with Applications, Volume 207, 2022, 117986, ISSN 0957-4174, https://doi. org/10. 1016/j. eswa. 2022. 117986. 

[25] Mo Yang, Jing Wang, Adaptability of Financial Time Series Prediction Based on BiLSTM, Procedia Computer Science, Volume 199, 2022, Pages 18-25, ISSN 1877-0509, https://doi. org/10. 1016/ j. procs. 2022. 01. 003. 

[26] Furkan Elmaz, Reinout Eyckerman, Wim Casteels, Steven Latré, Peter Hellinckx, CNN-LST-M architecture for predictive indoor temperature modeling, Building and Environment, Volume 206, 2021, 108327, ISSN 0360-1323, https://doi. org/10. 1016/j. buildenv. 2021. 108327. 

[27] Reuben Varghese Joseph, Anshuman Mohanty, Soumyae Tyagi, Shruti Mishra, Sandeep Ku-mar Satapathy, Sachi Nandan Mohanty, A hybrid deep learning framework with CNN and Bi-directional LSTM for store item demand forecasting, Computers and Electrical Engineering, Volume 103, 2022, 108358, ISSN 0045-7906, https://doi. org/10. 1016/j. compeleceng. 2022. 108358. 

[28] Huang N. E. , Shen Z. , Long S. , Wu M. , Shih H. , Zheng Q. , Liu H. (1998). Theem-pirical mode decomposition and the Hilbert spectrum for nonlinear and nonstationary time series analysis. Proceedings of the Royal SocietyA-Mathematical Physical and Engineering Sciences, 454(1971), 903–995. https://doi. org/10. 1098/ rspa. 1998. 0193. 

[29] Wu Z., & Huang N. E. (2009). Ensemble empirical mode decomposition: A noise-assisted data analysis method. Adv. Adapt. Data Anal. , 1, 1–41. https://doi. org/10. 1142/ S1793536909000047.