New Approaches to Forecasting Budget Revenues of the Russian Federation Based on Reservoir Computing

The relevance of this study stems from the need to enhance the accuracy of forecasting tools in determining future budget revenues for the Russian Federation, given the dynamic macroeconomic environment shaped by sanctions. In the current situation, it is essential to respond quickly to the changes taking place. This requires the use of various frequency data in predictive models and the search for new, more accurate forecasting methods. The object of the study is the dynamics of federal budget revenues. The subject of the research is to examine the applicability of reservoir computing in forecasting federal budget revenues in the Russian Federation. The purpose of the study is to identify the feasibility of using reservoir computing models in forecasting federal budget revenues in the Russian Federation. Empirical and theoretical methods were employed in the research process. These methods allowed us to understand the essence of reservoir computing, interpret the predictive results obtained, and select the best hyperparameters. As a result, a model based on reservoir computing was proposed by the author, taking into account the dynamics of monthly and daily factors in the development of the Russian economy. It is concluded that the world’s first experience in using reservoir computing in forecasting federal budget revenues in the Russian Federation has improved the quality of the model. The characteristics of the resulting model are significantly better than analogues calculated using other methods. The high fragmentation of the Russian data and the short length of the time series have also been revealed, which was eliminated by shortening the time period for training models and imputing missing values in the data.

1. Glenday G. Revenue forecasting. In: Allen R., Hemming R., Potter B. H., eds. The international handbook of public financial management. London: Palgrave Macmillan; 2013:435-452. DOI: https://doi.org/10.1057/9781137315304_21

2. Noor N., Sarlan A., Aziz N. Government revenue prediction using feed forward neural network. Journal of Theoretical and Applied Information Technology. 2023;101(6):2459-2473.

3. Batóg B., Batóg J. Regional government revenue forecasting: Risk factors of investment financing. Risks. 2021;9(12):210. DOI: 10.3390/risks9120210

4. Serkov L. A. Critical approach to the analysis of the problems of dynamic stochastic general equilibrium model. Ekonomika i biznes: teoriya i praktika = Economy and Business: Theory and Practice. 2015;(8):122-126. (In Russ.).

5. Storm S. Cordon of conformity: Why DSGE models are not the future of macroeconomics. International Journal of Political Economy. 2021;50(2):77-98. DOI: 10.1080/08911916.2021.1929582

6. Abdi G., Mazur T., Szaciłowski K. An organized view of reservoir computing: A perspective on theory and technology development. Japanese Journal of Applied Physics. 2024;63(5):050803. DOI: 10.35848/1347-4065/ad394f

7. Karaev A. K., Borisova O. V. Prospective models of financial forecasting of budget revenues. Finance: Theory and Practice. 2025;29(1):20-33. DOI: 10.26794/2587-5671-2025-29-1-20-33

8. Ballarin G., et al. Reservoir computing for macroeconomic forecasting with mixed-frequency data. International Journal of Forecasting. 2024;40(3):1206-1237. DOI: 10.1016/j.ijforecast.2023.10.009

9. Santos A., Lima R. R., Alves J. L., Misturini D. W., Florindo J. B. Reservoir computing and non-linear dynamics for time series analysis: An application in the financial market. Physica D: Nonlinear Phenomena. 2025;476:134698. DOI: 10.1016/j.physd.2025.134698

10. Jaeger H. The “echo state” approach to analysing and training recurrent neural networks — with an erratum note. German National Research Center for Information Technology. GMD Report. 2001;(148). URL: https://www.researchgate.net/publication/215385037_The_echo_state_approach_to_analysing_and_training_recurrent_neural_networks-with_an_erratum_note’

11. Jaeger H., Haas H. Harnessing nonlinearity: Predicting chaotic systems and saving energy in wireless communication. Science. 2004;304(5667):78-80. DOI: 10.1126/science.1091277

12. Pathak J., et al. Model-free prediction of large spatiotemporally chaotic systems from data: A reservoir computing approach. Physical Review Letters. 2018;120:024102. DOI: 10.1103/PhysRevLett.120.024102

13. Gauthier D. J., et al. Next generation reservoir computing. Nature Communications. 2021;12:5564. DOI: 10.1038/s41467-021-25801-2

14. Platt J. A., Wong A., Clark R., Penny S. G., Abarbanel H. D.I. Forecasting using reservoir computing: The role of generalized synchronization. arXiv preprint. 2021. DOI: 10.48550/arXiv.2103.00362

15. Tziatzios P. Financial nonlinear time-series analysis and prediction with reservoir computing. Master of science thesis in data science. School of Science & Technology. 2019.

16. Bollt E. M. On explaining the surprising success of reservoir computing forecaster of chaos? The universal machine learning dynamical system with contrast to VAR and DMD. Chaos: An Interdisciplinary Journal of Nonlinear Science. 2021;31(1):013108. DOI: 10.1063/5.0024890

17. Borisova O. V. Separate aspects of financial forecasting in the public sector. RISK: resursy, informatsiya, snabzhenie, konkurentsiya = RISK: Resources, Information, Supply, Competition. 2023;(2):177-181. (In Russ.). DOI: 10.56584/1560-8816-2023-2-177-181

18. Karaev A. K., Ponkratov V. V. Basics of the Wolfgang Stützel’s balance mechanics. Mir novoi ekonomiki = The World of New Economy. 2018;12(1):104-113. (In Russ.). DOI: 10.26794/2220-6469-2018-12-1-104-113

19. Karaev A. K., Ponkratov V. V. The relationship between financial development and economic growth in Russia (2000-2022). Ekonomika. Nalogi. Pravo = Economics, Taxes & Law. 2024;17(2):17-26. (In Russ.). DOI: 10.26794/1999-849X-2024-17-2-17-26