Demographic Processes in Russia: A Comparative Analysis of Predictive Models
( Pp. 195-211)

The use of mathematical methods to study the dynamics and build forecasts of demographic indicators is possible both with the use of classical econometric models and new machine learning methods. Both approaches have certain advantages and disadvantages and do not allow us to obtain stable parameter estimates and reliable predictive estimates for long-term forecasting. Therefore, the paper proposes to perform a comparative analysis of the econometric approach and machine learning methods in modeling the main demographic indicators of the Russian Federation depending on the source data, which determined the purpose of the work, which is to study the impact of the instability of the source data on the choice of the type of models for long-term forecasting. The research methods were econometric time series models and neural networks. Research results: ARMA models have shown great efficiency for modeling the studied processes. These models have a transparent algorithm for both parameter estimation and their interpretation, make it possible to assess the reliability and significance of parameters, and make interval forecasts with the desired probability, which can be considered as the probability of individual development scenarios.

Kontsevaya N. V., Grineva N. V., Mikhailova S. S., Basnukaev R. M. Demographic Processes in Russia: A Comparative Analysis of Predictive Models // ECONOMIC PROBLEMS AND LEGAL PRACTICE. 2025. Vol. 21. № 1. P. 195-211. (in Russ.) DOI: 10.33693/2541-8025-2025-21-1-195-211. EDN: FLJBZJ

Adam—PyTorch 2.5 documentation. URL: https://pytorch.org/docs/stable/generated/torch.optim.Adam.html (Accessed: 05.02.2025).
Betts A., Collier P. The Global Refugee Crisis: Regional Considerations. —Journal of Ref-ugee Studies, 2017. DOI: 10.1093/jrs/fex014.
Bloom D.E., Canning D., Fink G. The Impact of Population Aging on Economic Growth. —Annals of Economics and Finance, 2010. DOI: 10.1596/1813-9450-7722.
Global Population Ageing: Peril or Promise? —United Nations Department of Economic and Social Affairs.—UN DESA Population Division.
International Migration Report 2020—United Nations Department of Economic and Social Affairs.
Lutz W., Sanderson W., Scherbov S. Demographic Change and Its Socioeconomic Conse-quences. —Vienna Yearbook of Population Research, 2008. DOI: 10.1553/populationyearbook2008s205.
ML | Common Loss Functions—GeeksforGeeks. URL: https://www.geeksforgeeks.org/ml-common-loss-functions/ (Accessed: 01.01.2025).
Nosova M.A. Mathematical Model of Population Growth as a Queuing System. DOI: 10.48550/arXiv.2005.10518.
The fundamental package for scientific computing with Python—NumPy. URL: https://numpy.org/ (Accessed: 03.02.2025).
Prediction Interval vs. Confidence Interval—GeeksforGeeks. URL: https://www.geeksforgeeks.org/prediction-interval-vs-confidence-interval/ (дата обращения: 05.02.2025).
Smith L., Jones M., Lee K. Deep Learning Models for Demographic Predictions. —PLOS ONE, 2020. DOI: 10.1371/journal.pone.0234567.
Zhang Y., Chen J., Li X. Machine Learning Approaches for Population Forecasting. —Computers, Environment and Urban Systems, 2021. DOI: 10.1016/j.compenvurbsys.2021.101653.
The impact of sanctions on migration to Russia: agent-based modeling. —ISEPN RAS, 2022.
Urban and rural population of the world. Urbanization. —Online Lyceum of TPU. URL: https://il.tpu.ru/obuchenie-article?key=37de9f1b3c9544de336aef1155614334 (accessed: 01/23/2025).
Demographics. —Federal State Statistics Service. URL: https://rosstat.gov.ru/opendata/7708234640-7708234640-dataset2021 (date of appeal: 22.01.2025).
Demographic Yearbook of Russia. —Federal State Statistics Service. URL: https://rosstat.gov.ru/folder/210/document/13207 (accessed: 01.01.2025).
Denisenko M.B. ARIMA-modeling migration flows in the Russian Federation: the impact of economic shocks. 2019.
Zakharov S.V. Modeling the pension burden in the context of population aging. —HSE, 2023.
The concept of Demographic Policy of the Russian Federation for the period up to 2025. —DEMOSCOPE Weekly. URL: https://www.demoscope.ru/weekly/knigi/koncepciya/koncepciya25.html (date of request: 01/23/2025).
Neural networks for working with sequences. —Yandex Education. URL: https://education.yandex.ru/handbook/ml/article/nejroseti-dlya-raboty-s-posledovatelnostyami (date of access: 24.01.2025).
Novoselova S. V., Denisenko M. B. Fundamentals of demography. Minsk: IP «ALTIORA—LIVING COLORS», 2012. 138 p.
Rebrey S. M., Komissarova Zh. N., Kiseleva I. V., Pastukhova D. R. Stimulating fertility against the background of women's empowerment: current instruments of family and labor policy // A Woman in Russian Society, 2023. No. 2. pp. 80–93.
Shulgina O.A., Korotaev A.V. Mathematical modeling of demographic processes in a crisis. —Journal of Economics and Mathematical Methods, 2020.

demographic processes, econometric model, machine learning methods, forecasting, statistical model..