Deep learning approaches for MIMO timeseries analysis 
^{(2) } Sarina Sulaiman (Soft Computing Research Group, Faculty of Computing, Universiti Teknologi Malaysia, Johor Bahru, Malaysia)
^{(3) } Siaka Konate (Department of Electronic and Telecommunications, Normal School of Technical and Vocational Education, Bamoko, Mali)
^{(4) } Modawy Adam Ali Abdalla (College of Energy and Electrical Engineering, Hohai University, Nanjing, China; and Department of Electrical and Electronic Engineering, College of Engineering Science, Nyala University, Nyala,, Sudan)
^{*}corresponding author
AbstractThis study presents a comparative analysis of various deep learning (DL) methods for multiinput and multioutput (MIMO) timeseries forecasting of stock prices. The analysis is conducted on a dataset comprising the stock price of Bitcoin. The dataset consists of 2950 rows from December 2017 to December 2021. This study aims to evaluate the performance of multiple DL methods, including Multilayer Perceptron (MLP), Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), Long Short Term Memory (LSTM), Bidirectional LSTM (BiLSTM), and Gated Recurrent Unit (GRU). The evaluation criteria for selecting the bestperforming methods in this research are based on two performance metrics: Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE). These metrics were chosen for specific reasons related to assessing the accuracy and reliability of the forecasting models. MAPE is used to assess accuracy, while RMSE helps detect outliers in the system. Results show that the LSTM method achieves the best performance, outperforming other methods with an average MAPE value of 8.73% and BiLSTM has the best average RMSE value of 0.02216. The findings of this study have practical implications for timeseries forecasting in the field of stock trading. The superior performance of LSTM highlights its potential as a reliable method for accurately predicting stock prices. The BiLSTM model's ability to detect outliers can aid in identifying abnormal stock market behavior. In summary, this research provides insights into the performance of various DL models of MIMO for stock price forecasting. The results contribute to the field of timeseries forecasting and offer valuable guidance for decisionmaking in stock trading by identifying the most effective methods for predicting stock prices accurately and detecting unusual market behavior.
KeywordsMIMO; Time series; Deep learning; forecasting

DOIhttps://doi.org/10.26555/ijain.v9i2.1092 
Article metricsAbstract views : 754  PDF views : 205 
Cite 
Full TextDownload 
References
[1] O. B. Sezer, M. U. Gudelek, and A. M. Ozbayoglu, “Financial time series forecasting with deep learning : A systematic literature review: 2005–2019,” Appl. Soft Comput., vol. 90, p. 106181, May 2020, doi: 10.1016/j.asoc.2020.106181.
[2] M. S. Gorus and M. Aydin, “The relationship between energy consumption, economic growth, and CO2 emission in MENA countries: Causality analysis in the frequency domain,” Energy, vol. 168, pp. 815–822, Feb. 2019, doi: 10.1016/j.energy.2018.11.139.
[3] H. Lan, C. Zhang, Y.Y. Hong, Y. He, and S. Wen, “Dayahead spatiotemporal solar irradiation forecasting using frequencybased hybrid principal component analysis and neural network,” Appl. Energy, vol. 247, pp. 389–402, Aug. 2019, doi: 10.1016/j.apenergy.2019.04.056.
[4] K. Bandara, P. Shi, C. Bergmeir, H. Hewamalage, Q. Tran, and B. Seaman, “Sales Demand Forecast in Ecommerce Using a Long ShortTerm Memory Neural Network Methodology,” in Lecture Notes in Computer Science, 2019, pp. 462–474, doi: 10.1007/9783030367183_39.
[5] T. Bikku, “Multilayered deep learning perceptron approach for health risk prediction,” J. Big Data, vol. 7, no. 1, p. 50, Dec. 2020, doi: 10.1186/s40537020003167.
[6] P. Hewage, M. Trovati, E. Pereira, and A. Behera, “Deep learningbased effective finegrained weather forecasting model,” Pattern Anal. Appl., vol. 24, no. 1, pp. 343–366, Feb. 2021, doi: 10.1007/s10044020008981.
[7] X. Yang, X. Liu, and Z. Li, “Multimodel Approach to Robust Identification of MultipleInput SingleOutput Nonlinear TimeDelay Systems,” IEEE Trans. Ind. Informatics, vol. 16, no. 4, pp. 2413–2422, Apr. 2020, doi: 10.1109/TII.2019.2933030.
[8] C. Deng, Y. Huang, N. Hasan, and Y. Bao, “Multistepahead stock price index forecasting using long shortterm memory model with multivariate empirical mode decomposition,” Inf. Sci. (Ny)., vol. 607, pp. 297–321, Aug. 2022, doi: 10.1016/j.ins.2022.05.088.
[9] K. Li, G. Huang, S. Wang, B. Baetz, and W. Xu, “A Stepwise Clustered Hydrological Model for Addressing the Temporal Autocorrelation of Daily Streamflows in Irrigated Watersheds,” Water Resour. Res., vol. 58, no. 2, pp. 1–31, Feb. 2022, doi: 10.1029/2021WR031065.
[10] F. Amato, F. Guignard, S. Robert, and M. Kanevski, “A novel framework for spatiotemporal prediction of environmental data using deep learning,” Sci. Rep., vol. 10, no. 1, p. 22243, Dec. 2020, doi: 10.1038/s41598020791487.
[11] Z. Qu et al., “Temperature forecasting of grain in storage: A multioutput and spatiotemporal approach based on deep learning,” Comput. Electron. Agric., vol. 208, p. 107785, May 2023, doi: 10.1016/j.compag.2023.107785.
[12] R. Rakholia, Q. Le, B. Quoc Ho, K. Vu, and R. Simon Carbajo, “Multioutput machine learning model for regional air pollution forecasting in Ho Chi Minh City, Vietnam,” Environ. Int., vol. 173, p. 107848, Mar. 2023, doi: 10.1016/j.envint.2023.107848.
[13] A. Thakkar and K. Chaudhari, “Predicting stock trend using an integrated term frequency–inverse document frequencybased feature weight matrix with neural networks,” Appl. Soft Comput., vol. 96, p. 106684, Nov. 2020, doi: 10.1016/j.asoc.2020.106684.
[14] G. Ding and L. Qin, “Study on the prediction of stock price based on the associated network model of LSTM,” Int. J. Mach. Learn. Cybern., vol. 11, no. 6, pp. 1307–1317, Jun. 2020, doi: 10.1007/s13042019010411.
[15] J. Silva et al., “An Early Warning Method for Agricultural Products Price Spike Based on Artificial Neural Networks Prediction,” in Lecture Notes in Computer Science, 2019, pp. 622–632, doi: 10.1007/9783030227418_44.
[16] S. Mishra, C. Bordin, K. Taharaguchi, and I. Palu, “Comparison of deep learning models for multivariate prediction of time series wind power generation and temperature,” Energy Reports, vol. 6, pp. 273–286, Feb. 2020, doi: 10.1016/j.egyr.2019.11.009.
[17] C. V. Hudiyanti, J. L. Buliali, and A. Saikhu, “Modelling MIMO Transfer Functions for Analysis of The Relationship Between Temperature and Air Humidity with the Number of Confirmation, Suspect and Probable COVID19 in Surabaya,” in 2021 International Conference on Artificial Intelligence and Computer Science Technology (ICAICST), Jun. 2021, pp. 246–251, doi: 10.1109/ICAICST53116.2021.9497836.
[18] H. Rodriguez, M. Medrano, L. M. Rosales, G. P. Penunuri, and J. J. Flores, “Multistep forecasting strategies for wind speed time series,” in 2020 IEEE International Autumn Meeting on Power, Electronics and Computing (ROPEC), Nov. 2020, pp. 1–6, doi: 10.1109/ROPEC50909.2020.9258743.
[19] P. Hewage et al., “Temporal convolutional neural (TCN) network for an effective weather forecasting using timeseries data from the local weather station,” Soft Comput., vol. 24, no. 21, pp. 16453–16482, Nov. 2020, doi: 10.1007/s00500020049540.
[20] A. Alzaghir, A. R. Abdellah, and A. Koucheryavy, “Predicting Energy Consumption for UAVEnabled MEC Using Machine Learning Algorithm,” in Lecture Notes in Computer Science, 2022, pp. 297–309, doi: 10.1007/9783030977771_25.
[21] D. Saxena, I. Gupta, A. K. Singh, and C.N. Lee, “A Fault Tolerant Elastic Resource Management Framework Toward High Availability of Cloud Services,” IEEE Trans. Netw. Serv. Manag., vol. 19, no. 3, pp. 3048–3061, Sep. 2022, doi: 10.1109/TNSM.2022.3170379.
[22] J.J. Zhang and H.S. Yan, “MTN optimal control of MIMO nonaffine nonlinear timevarying discrete systems for tracking only by output feedback,” J. Franklin Inst., vol. 356, no. 8, pp. 4304–4334, May 2019, doi: 10.1016/j.jfranklin.2019.03.008.
[23] T. Niu, J. Wang, H. Lu, and P. Du, “Uncertainty modeling for chaotic time series based on optimal multiinput multioutput architecture: Application to offshore wind speed,” Energy Convers. Manag., vol. 156, pp. 597–617, Jan. 2018, doi: 10.1016/j.enconman.2017.11.071.
[24] P. LaraBenítez, L. GallegoLedesma, M. CarranzaGarcía, and J. M. LunaRomera, “Evaluation of the Transformer Architecture for Univariate Time Series Forecasting,” in Lecture Notes in Computer Science, 2021, pp. 106–115, doi: 10.1007/9783030857134_11.
[25] C. Yin and Q. Dai, “A deep multivariate time series multistep forecasting network,” Appl. Intell., vol. 52, no. 8, pp. 8956–8974, Jun. 2022, doi: 10.1007/s1048902102899x.
[26] Z. Wu, G. Luo, Z. Yang, Y. Guo, K. Li, and Y. Xue, “A comprehensive review on deep learning approaches in wind forecasting applications,” CAAI Trans. Intell. Technol., vol. 7, no. 2, pp. 129–143, Jun. 2022, doi: 10.1049/cit2.12076.
[27] H. Huang, Y. Wang, Y. Li, Y. Zhou, and Z. Zeng, “DebrisFlow Susceptibility Assessment in China: A Comparison between Traditional Statistical and Machine Learning Methods,” Remote Sens., vol. 14, no. 18, p. 4475, Sep. 2022, doi: 10.3390/rs14184475.
[28] A. Truchot et al., “Machine learning does not outperform traditional statistical modelling for kidney allograft failure prediction,” Kidney Int., vol. 103, no. 5, pp. 936–948, May 2023, doi: 10.1016/j.kint.2022.12.011.
[29] M. Sousa, A. M. Tomé, and J. Moreira, “Longterm forecasting of hourly retail customer flow on intermittent time series with multiple seasonality,” Data Sci. Manag., vol. 5, no. 3, pp. 137–148, Sep. 2022, doi: 10.1016/j.dsm.2022.07.002.
[30] A. L. Schaffer, T. A. Dobbins, and S.A. Pearson, “Interrupted time series analysis using autoregressive integrated moving average (ARIMA) models: a guide for evaluating largescale health interventions,” BMC Med. Res. Methodol., vol. 21, no. 1, p. 58, Dec. 2021, doi: 10.1186/s12874021012358.
[31] H. Taud and J. F. Mas, “Multilayer Perceptron (MLP),” in Lecture Notes in Geoinformation and Cartography, 2018, pp. 451–455, doi: 10.1007/9783319608013_27.
[32] B. Cai et al., “Resilience evaluation methodology of engineering systems with dynamicBayesiannetworkbased degradation and maintenance,” Reliab. Eng. Syst. Saf., vol. 209, p. 107464, May 2021, doi: 10.1016/j.ress.2021.107464.
[33] R. Reichenberg, “Dynamic Bayesian Networks in Educational Measurement: Reviewing and Advancing the State of the Field,” Appl. Meas. Educ., vol. 31, no. 4, pp. 335–350, Oct. 2018, doi: 10.1080/08957347.2018.1495217.
[34] A. Safari and M. Davallou, “Oil price forecasting using a hybrid model,” Energy, vol. 148, pp. 49–58, Apr. 2018, doi: 10.1016/j.energy.2018.01.007.
[35] T. Kattenborn, J. Leitloff, F. Schiefer, and S. Hinz, “Review on Convolutional Neural Networks (CNN) in vegetation remote sensing,” ISPRS J. Photogramm. Remote Sens., vol. 173, pp. 24–49, Mar. 2021, doi: 10.1016/j.isprsjprs.2020.12.010.
[36] J. Lu, L. Tan, and H. Jiang, “Review on Convolutional Neural Network (CNN) Applied to Plant Leaf Disease Classification,” Agriculture, vol. 11, no. 8, p. 707, Jul. 2021, doi: 10.3390/agriculture11080707.
[37] R. Chauhan, K. K. Ghanshala, and R. . Joshi, “Convolutional Neural Network (CNN) for Image Detection and Recognition,” in 2018 First International Conference on Secure Cyber Computing and Communication (ICSCCC), Dec. 2018, pp. 278–282, doi: 10.1109/ICSCCC.2018.8703316.
[38] I. E. Livieris, N. Kiriakidou, S. Stavroyiannis, and P. Pintelas, “An Advanced CNNLSTM Model for Cryptocurrency Forecasting,” Electronics, vol. 10, no. 3, p. 287, Jan. 2021, doi: 10.3390/electronics10030287.
[39] T. Fischer and C. Krauss, “Deep learning with long shortterm memory networks for financial market predictions,” Eur. J. Oper. Res., vol. 270, no. 2, pp. 654–669, Oct. 2018, doi: 10.1016/j.ejor.2017.11.054.
[40] S. Ghimire, Z. M. Yaseen, A. A. Farooque, R. C. Deo, J. Zhang, and X. Tao, “Streamflow prediction using an integrated methodology based on convolutional neural network and long shortterm memory networks,” Sci. Rep., vol. 11, no. 1, p. 17497, Sep. 2021, doi: 10.1038/s41598021967514.
[41] N. Q. K. Le, E. K. Y. Yapp, and H.Y. Yeh, “ETGRU: using multilayer gated recurrent units to identify electron transport proteins,” BMC Bioinformatics, vol. 20, no. 1, p. 377, Dec. 2019, doi: 10.1186/s1285901929725.
[42] A.N. Buturache and S. Stancu, “Solar Energy Production Forecast Using Standard Recurrent Neural Networks, Long ShortTerm Memory, and Gated Recurrent Unit,” Eng. Econ., vol. 32, no. 4, pp. 313–324, Oct. 2021, doi: 10.5755/j01.ee.32.4.28459.
[43] H. V. Bitencourt, O. Orang, L. A. F. de Souza, P. C. L. Silva, and F. G. Guimarães, “An embeddingbased nonstationary fuzzy time series method for multiple output highdimensional multivariate time series forecasting in IoT applications,” Neural Comput. Appl., vol. 35, no. 13, pp. 9407–9420, May 2023, doi: 10.1007/s00521022081205.
[44] U. Kamath, J. Liu, and J. Whitaker, “Convolutional Neural Networks,” in Deep Learning for NLP and Speech Recognition, Cham: Springer International Publishing, 2019, pp. 263–314, doi: 10.1007/9783030145965_6.
[45] K. Sekaran, P. Chandana, N. M. Krishna, and S. Kadry, “Deep learning convolutional neural network (CNN) With Gaussian mixture model for predicting pancreatic cancer,” Multimed. Tools Appl., vol. 79, no. 15–16, pp. 10233–10247, Apr. 2020, doi: 10.1007/s1104201974195.
[46] S. Fan, N. Xiao, and S. Dong, “A novel model to predict significant wave height based on long shortterm memory network,” Ocean Eng., vol. 205, p. 107298, Jun. 2020, doi: 10.1016/j.oceaneng.2020.107298.
[47] G. Kumar, S. Jain, and U. P. Singh, “Stock Market Forecasting Using Computational Intelligence: A Survey,” Arch. Comput. Methods Eng., vol. 28, no. 3, pp. 1069–1101, May 2021, doi: 10.1007/s11831020094135.
[48] S. Carta, A. Ferreira, A. S. Podda, D. Reforgiato Recupero, and A. Sanna, “MultiDQN: An ensemble of Deep Qlearning agents for stock market forecasting,” Expert Syst. Appl., vol. 164, p. 113820, Feb. 2021, doi: 10.1016/j.eswa.2020.113820.
[49] Y. Kumar, A. Koul, S. Kaur, and Y.C. Hu, “Machine Learning and Deep Learning Based Time Series Prediction and Forecasting of Ten Nations’ COVID19 Pandemic,” SN Comput. Sci., vol. 4, no. 1, p. 91, Dec. 2022, doi: 10.1007/s42979022014933.
[50] A. P. Wibawa, I. T. Saputra, A. B. P. Utama, W. Lestari, and Z. N. Izdihar, “Long ShortTerm Memory to Predict Unique Visitors of an Electronic Journal,” in 2020 6th International Conference on Science in Information Technology (ICSITech), Oct. 2020, pp. 176–179, doi: 10.1109/ICSITech49800.2020.9392031.
[51] A. P. Wibawa, A. B. P. Utama, H. Elmunsyah, U. Pujianto, F. A. Dwiyanto, and L. Hernandez, “Timeseries analysis with smoothed Convolutional Neural Network,” J. Big Data, vol. 9, no. 1, p. 44, Dec. 2022, doi: 10.1186/s4053702200599y.
[52] A. R. F. Dewandra, A. P. Wibawa, U. Pujianto, A. B. P. Utama, and A. Nafalski, “Journal Unique Visitors Forecasting Based on Multivariate Attributes Using CNN,” Int. J. Artif. Intell. Res., vol. 6, no. 2, pp. 18, Jul. 2022, doi: 10.29099/ijair.v6i1.274.
[53] P. Dhruv and S. Naskar, “Image Classification Using Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN): A Review,” in Advances in Intelligent Systems and Computing, 2020, pp. 367–381, doi: 10.1007/9789811518843_34.
[54] A. Sherstinsky, “Fundamentals of Recurrent Neural Network (RNN) and Long ShortTerm Memory (LSTM) network,” Phys. D Nonlinear Phenom., vol. 404, p. 132306, Mar. 2020, doi: 10.1016/j.physd.2019.132306.
[55] A. W. Saputra, A. P. Wibawa, U. Pujianto, A. B. P. Utama, and A. Nafalski, “LSTMbased Multivariate TimeSeries Analysis : A Case of Journal Visitors Forecasting,” Ilk. J. Ilm., vol. 14, no. 1, pp. 57–62, 2022, doi: 10.33096/ilkom.v14i1.1106.5762.
[56] A. P. Wibawa, R. R. Ula, A. B. P. Utama, M. Y. Chuttur, A. Pranolo, and Haviluddin, “Forecasting eJournal Unique Visitors using Smoothed Long ShortTerm Memory,” in 2021 7th International Conference on Electrical, Electronics and Information Engineering (ICEEIE), Oct. 2021, pp. 609–613, doi: 10.1109/ICEEIE52663.2021.9616628.
[57] A. Pranolo, Y. Mao, A. P. Wibawa, A. B. P. Utama, and F. A. Dwiyanto, “Robust LSTM With TunedPSO and BifoldAttention Mechanism for Analyzing Multivariate TimeSeries,” IEEE Access, vol. 10, pp. 78423–78434, 2022, doi: 10.1109/ACCESS.2022.3193643.
[58] P. L. Seabe, C. R. B. Moutsinga, and E. Pindza, “Forecasting Cryptocurrency Prices Using LSTM, GRU, and BiDirectional LSTM: A Deep Learning Approach,” Fractal Fract., vol. 7, no. 2, p. 203, Feb. 2023, doi: 10.3390/fractalfract7020203.
[59] A. Zeroual, F. Harrou, A. Dairi, and Y. Sun, “Deep learning methods for forecasting COVID19 timeSeries data: A Comparative study,” Chaos, Solitons & Fractals, vol. 140, p. 110121, Nov. 2020, doi: 10.1016/j.chaos.2020.110121.
[60] G. Yigit and M. F. Amasyali, “Simple But Effective GRU Variants,” in 2021 International Conference on INnovations in Intelligent SysTems and Applications (INISTA), Aug. 2021, pp. 1–6, doi: 10.1109/INISTA52262.2021.9548535.
[61] J. Zhao, H. Qu, J. Zhao, H. Dai, and D. Jiang, “Spatiotemporal graph convolutional recurrent networks for traffic matrix prediction,” Trans. Emerg. Telecommun. Technol., vol. 31, no. 11, pp. 114, Nov. 2020, doi: 10.1002/ett.4056.
[62] A. B. P. Utama, A. P. Wibawa, Muladi, and A. Nafalski, “PSO based Hyperparameter tuning of CNN Multivariate TimeSeries Analysis,” J. Online Inform., vol. 7, no. 2, pp. 193–202, 2022, doi: 10.15575/join.v7i2.858.
[63] Y. Mao, A. Pranolo, A. P. Wibawa, A. B. Putra Utama, F. A. Dwiyanto, and S. Saifullah, “Selection of Precise Long Short Term Memory (LSTM) Hyperparameters based on Particle Swarm Optimization,” in 2022 International Conference on Applied Artificial Intelligence and Computing (ICAAIC), May 2022, pp. 1114–1121, doi: 10.1109/ICAAIC53929.2022.9792708.
[64] A. P. Wibawa, Z. N. Izdihar, A. B. P. Utama, L. Hernandez, and Haviluddin, “MinMax Backpropagation Neural Network to Forecast eJournal Visitors,” in 2021 International Conference on Artificial Intelligence in Information and Communication (ICAIIC), Apr. 2021, pp. 052–058, doi: 10.1109/ICAIIC51459.2021.9415197.
[65] M. Alhussein, K. Aurangzeb, and S. I. Haider, “Hybrid CNNLSTM Model for ShortTerm Individual Household Load Forecasting,” IEEE Access, vol. 8, pp. 180544–180557, 2020, doi: 10.1109/ACCESS.2020.3028281.
This work is licensed under a Creative Commons AttributionShareAlike 4.0 International License.
___________________________________________________________
International Journal of Advances in Intelligent Informatics
ISSN 24426571 (print)  25483161 (online)
Organized by UAD and ASCEE Computer Society
Published by Universitas Ahmad Dahlan
W: http://ijain.org
E: info@ijain.org (paper handling issues)
andri.pranolo.id@ieee.org (publication issues)
This work is licensed under a Creative Commons AttributionShareAlike 4.0