1. SISN
  2. All Volumes and Issues
  3. Volume 19, 2026
  4. Time Series Forecasting Methods

Time Series Forecasting Methods

SISN.
2026;
Volume 19
: pp. 382 - 392
ISSN: 2524-065Х (рrint); 2663-0001 (оnline)

https://doi.org/10.23939/sisn2026.19.382
Received: March 06, 2026
Accepted: April 22, 2026
Authors:
  1. Dmytro Nakonechnyi
  2. Iryna Zavushchak
1
Lviv Polytechnic National University, Information Systems and Networks Department, Lviv, Ukraine
ORCID: 0009-0001-7763-7275
2
Lviv Polytechnic National University, Information Systems and Networks Department, Lviv, Ukraine
ORCID: 0000-0002-5371-8775

The article investigates the limitations of modern approaches to time series forecasting in complex dynamic and nonlinear processes whose structure may consist of heterogeneous data types. The relevance of the study is driven by the rapid growth of data volumes in information systems, their diversity, and the need to improve forecasting accuracy under conditions of non-stationarity and multifactor influence.

The aim of the research is to analyze modern time series forecasting methods, identify their limitations, and generalize these limitations in the form of a classification, as well as to provide their formalized comparison.

The study employs comparative analysis methods for classical statistical models, machine learning algorithms, deep neural networks, and hybrid approaches based on criteria such as the ability to model nonlinear dependencies, adaptability to non-stationarity, integration of heterogeneous data, scalability, and computational complexity.

As a result, a classification of limitations of time series forecasting methods is proposed, including data-related limitations, model limitations, computational limitations, and system-level limitations. In addition, a formalized comparison of different classes of methods according to the defined criteria is conducted.

It is shown that, in the general case of complex multimodal and non-stationary processes, none of the considered classes of methods ensures the simultaneous achievement of high accuracy, adaptability, scalability, and effective integration of heterogeneous data.

The practical significance of the study lies in the possibility of using the obtained results to justify the need for developing new forecasting approaches capable of comprehensively addressing the structural complexity and dynamic nature of data in modern information systems

time series
forecasting
Multimodal data
machine learning
deep neural networks
non-stationarity
  1. Balasubramaniam, T., Mohotti, W. A., Sabir, K., & Nayak, R. (2025). Feature engineering on climate data with machine learning to understand time-lagging effects in pasture yield prediction. Ecological Informatics, 86, 103011. https://doi.org/10.1016/j.ecoinf.2025.103011
  2. Bustarviejo, J., & Bousoño-Calzón, C. (2026). Multimodal information fusion for financial forecasting via cross-attention and calibrated uncertainty. Machine Learning with Applications, 23, 100840. https://doi.org/10.1016/j.mlwa.2026.100840
  3. Cabrera Martin, I., Mukherjee, S., Baimagambetov, A., Vanschoren, J., & Polatidis, N. (2025). Evolving machine learning in non-stationary environments: A unified survey of drift, forgetting, and adaptation. arXiv. https://doi.org/10.48550/arXiv.2505.17902
  4. Farahani, M. A., Kalach, F. E., Harper, A., McCormick, M. R., Harik, R., & Wuest, T. (2025). Time-series forecasting in smart manufacturing systems: An experimental evaluation of the state-of-the-art algorithms. Robotics and Computer-Integrated Manufacturing, 95, 103010. https://doi.org/10.1016/j.rcim.2025.103010
  5. Gasparin, A., Lukovic, S., & Alippi, C. (2021). Deep learning for time series forecasting: The electric load case. CAAI Transactions on Intelligence Technology. https://doi.org/10.1049/cit2.12060
  6. Hajirahimi, Z., & Khashei, M. (2019). Hybrid structures in time series modeling and forecasting: A review. Engineering Applications of Artificial Intelligence, 86, 83–106. https://doi.org/10.1016/j.engappai.2019.08.018
  7. Hall, T., & Rasheed, K. (2025). A Survey of Machine Learning Methods for Time Series Prediction. Applied Sciences, 15(11), 5957. https://doi.org/10.3390/app15115957
  8. Kim, J., Kim, H., Kim, H., Lee, D., & Yoon, S. (2025). A comprehensive survey of deep learning for time series forecasting: architectural diversity and open challenges. Artificial Intelligence Review, 58(7). https://doi.org/10.1007/s10462-025-11223-9
  9. Kontopoulou, V. I., Panagopoulos, A. D., Kakkos, I., & Matsopoulos, G. K. (2023). A Review of ARIMA vs. Machine Learning Approaches for Time Series Forecasting in Data Driven Networks. Future Internet, 15(8), 255. https://doi.org/10.3390/fi15080255
  10. Lim, B., & Zohren, S. (2021). Time Series Forecasting With Deep Learning: A Survey. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 379(2194), 20200209. https://doi.org/10.1098/rsta.2020.0209
  11. Ospina, R., Gondim, J. A. M., Leiva, V., & Castro, C. (2023). An Overview of Forecast Analysis with ARIMA Models during the COVID-19 Pandemic: Methodology and Case Study in Brazil. Mathematics, 11(14), 3069. https://doi.org/10.3390/math11143069
  12. Slimane, M., Bedioui, N., & Besbes, M. (2025). Assessing integration orders for SARIMA modeling A hypothesis testing approach with information criterion hyperparameter selection, case of predicting gas consumption in central Tunisia. Energy Strategy Reviews, 61, 101866. https://doi.org/10.1016/j.esr.2025.101866
  13. Suárez-Cetrulo, A. L., Quintana, D., & Cervantes, A. (2023). A survey on machine learning for recurring concept drifting data streams. Expert Systems with Applications, 213, 118934. https://doi.org/10.1016/j.eswa.2022.118934
  14. Wang, X., Wu, Y., Zou, W., & Zhao, X. (2025). Hybrid Time Series Forecasting for Real-Time Electricity Market Demand Using ARIMA-LSTM and Scalable Cloud-Native Architecture. Informatica, 49(3). https://doi.org/10.31449/inf.v49i3.9474
  15. Yang, Y., Zhang, J., & Wang, L. (2024). Semi-supervised prediction method for time series based on Monte Carlo and time fusion feature attention. Applied Soft Computing, 167, 112283. https://doi.org/10.1016/j.asoc.2024.112283
  16. Yu, H., Wang, Z., Xie, Y., & Wang, G. (2024). A multi-granularity hierarchical network for long- and short-term forecasting on multivariate time series data. Applied Soft Computing, 157, 111537–111537. https://doi.org/10.1016/j.asoc.2024.111537
  17. Zhao, S., Zhou, X., Jin, M., Hou, Z., Yang, C., Li, Z., Wen, Q., Wang, Y., Wen, Y., & Yuan, X. (2024). Rethinking self-supervised learning for time series forecasting: A temporal perspective. Knowledge-Based Systems, 305, 112652. https://doi.org/10.1016/j.knosys.2024.112652
  18. Zhu, W., Song, G., Wang, L., & Liu, S. (2024). AnchorGT: Efficient and flexible attention architecture for scalable graph transformers. arXiv. https://doi.org/10.48550/arXiv.2405.03481