+7 925 966 4690, 9am6pm (GMT+3), Monday – Friday
ИД «Финансы и кредит»

JOURNALS

  

FOR AUTHORS

  

SUBSCRIBE

    
Finance and Credit
 

Short-term forecasting of inflation, output of goods and services using machine learning

Vol. 31, Iss. 1, JANUARY 2025

Received: 17 October 2024

Accepted: 14 November 2024

Available online: 30 January 2025

Subject Heading: MONETARY ACCOMMODATION

JEL Classification: C52, C53, E31, E37, E52

Pages: 91-112

https://doi.org/10.24891/fc.31.1.91

Larisa N. DROBYSHEVSKAYA Kuban State University (KubSU), Krasnodar, Russian Federation
ld@seatrade.ru

https://orcid.org/0000-0002-2860-4629

Nikita A. DANKOV Kuban State University (KubSU), Krasnodar, Russian Federation
nikit.dankov@yandex.ru

https://orcid.org/0009-0006-1990-0983

Subject. The article addresses short-term forecasts of key economic indicators obtained using machine learning models that can be used as prerequisites in medium-term models used for stress testing, scenario analysis, and development of recommendations on monetary policy.
Objectives. The study aims at improving the accuracy of short-term forecasting of inflation, output of goods and services, based on the use of various models, including machine learning, and determining the most optimal one.
Methods. The study rests on systems approach, methods of statistical analysis, mathematical modeling, and forecasting.
Results. Based on the use of 12 models (econometric, state space, gradient boosting, neural networks), we obtained estimates of the quality of forecasts based on RMSE, MAE metrics for inflation, and the output of goods and services for various three- and six-month intervals, starting from the 2nd half of 2023 to the 1st half of 2024 (14 periods in total). A ranking of models by average forecast accuracy was compiled.
Conclusions. Machine learning models demonstrated their viability in short-term forecasting of inflation, output of goods and services for 3 and 6 months, and can be used for monetary regulation. To forecast inflation, it is advisable to use econometric methods such as AR(1), for more volatile indicators, gradient boosting (LightGBM, CatBoost, XGBoost) and some neural network approaches (TFT, RNN (LSTM)) turn out to be the most accurate.

Keywords: inflation, goods and services, short-term forecasting, econometric modeling, machine learning

References:

  1. Nelyubina A. [Forecasting Regional Indicators Based on the Quarterly Projection Model]. Den'gi i kredit = Russian Journal of Money and Finance, 2021, vol. 80, no. 2, pp. 50–75. (In Russ.)
  2. Kryzhanovskii O., Zykov A. [DEMUR: A Regional Semi-Structural Model of the Ural Macroregion]. Den'gi i kredit= Russian Journal of Money and Finance, 2022, vol. 81, no. 4, pp. 52–85. (In Russ.)
  3. Baibuza I. [Inflation Forecasting Using Machine Learning Methods. Russian Journal of Money and Finance]. Den'gi i kredit = Russian Journal of Money and Finance, 2018, vol. 77, no. 4, pp. 42–59. (In Russ.)
  4. Pavlov E. [Forecasting Inflation in Russia Using Neural Networks]. Den'gi i kredit = Russian Journal of Money and Finance, 2020, vol. 79, no. 1, pp. 57–73. (In Russ.)
  5. Bukina T.V., Kashin D.V. [Regional inflation forecasting: Econometric models versus machine learning methods?]. Ekonomicheskii zhurnal Vysshei shkoly ekonomiki = The HSE Economic Journal, 2024, vol. 28, no. 1, pp. 81–107. URL: Link (In Russ.)
  6. Grishchenko V., Krylov I. [New Approaches to Measuring, Analysing, and Forecasting Prices: A Review of the Bank of Russia, NES, and HSE University Workshop]. Den'gi i kredit = Russian Journal of Money and Finance, 2024, vol. 83, no. 2, pp. 92–111. (In Russ.)
  7. Maliar L., Maliar S., Winant P.  Deep Learning for Solving Dynamic Economic Models. Journal of Monetary Economics, 2021, vol. 122, pp. 76–101. URL: Link
  8. Dyer J., Cannon P., Farmer J.D., Schmon S.M.  Black-Box Bayesian Inference for Agent-Based Models. Journal of Economic Dynamics and Control, 2024, vol. 161, no. 104827. URL: Link
  9. Barkan O., Benchimol J., Caspi I. et al. Forecasting CPI Inflation Components with Hierarchical Recurrent Neural Networks. International Journal of Forecasting, 2023, vol. 39, iss. 3, pp. 1145–1162. URL: Link
  10. Medeiros M.C., Vasconcelos G.F., Veiga A., Zilberman E. Forecasting Inflation in a Data-Rich Environment: The Benefits of Machine Learning Methods. Journal of Business & Economic Statistics, 2021, vol. 39, iss. 1, pp. 98–119. URL: Link
  11. Molnar C. Interpretable Machine Learning: A Guide for Making Black Box Models Explainable. 2nd ed. Germany, Independently Published, 2022, 328 p.
  12. Latypov R., Akhmedova E., Postolit E., Mikitchuk M.D. [Bottom-up Inflation Forecasting Using Machine Learning Methods]. Den'gi i kredit = Russian Journal of Money and Finance, 2024, vol. 83, no. 3, pp. 23–44. (In Russ.)
  13. Isakov A., Latypov R., Repin A. et al. [Hard Numbers: Open Consumer Price Database]. Den'gi i kredit= Russian Journal of Money and Finance, 2021, vol. 80, no. 1, pp. 104–119. (In Russ.)
  14. Mogilat A., Kryzhanovskii O., Shuvalova Zh., Murashov Ya. [DYFARUS: Dynamic Factor Model to Forecast GDP by Output Using Input-Output Tables]. Den'gi i kredit = Russian Journal of Money and Finance, 2024, vol. 83, no. 2, pp. 3–25. (In Russ.)
  15. Zhemkov M. [Assessment of Monthly GDP Growth Using Temporal Disaggregation Methods]. Den'gi i kredit = Russian Journal of Money and Finance, 2022, vol. 81, no. 2, pp. 79–104. (In Russ.)
  16. Demeshev B.B., Malakhovskaya O.A. [Macroeconomic forecasting with a Litterman’s BVAR model]. Ekonomicheskii zhurnal Vysshei shkoly ekonomiki = HSE Economic Journal, 2016, vol. 20, no. 4, pp. 691–710. URL: Link (In Russ.)
  17. Lomivorotov R.V. [The use of Bayesian methods for the analysis of monetary policy in Russia]. Prikladnaya ekonometrika = Applied Econometrics, 2015, no. 2, pp. 41–63. URL: Link (In Russ.)
  18. Pestova A., Mamonov M. [Estimating the Influence of Different Shocks on Macroeconomic Indicators and Developing Conditional Forecasts on the Basis of BVAR Model for the Russian Economy]. Ekonomicheskaya politika = Economic Policy, 2016, vol. 11, no. 4, pp. 56–92. URL: Link (In Russ.)
  19. Korotkikh O. [A Multy-Country BVAR Model for the External Sector]. Den'gi i kredit = Russian Journal of Money and Finance, 2020, vol. 79, no. 4, pp. 98–112. (In Russ.)
  20. Shulyak E. [Macroeconomic Forecasting Using Data from Social Media]. Den'gi i kredit = Russian Journal of Money and Finance, 2022, vol. 81, no. 4, pp. 86–112. (In Russ.)

View all articles of issue

 

ISSN 2311-8709 (Online)
ISSN 2071-4688 (Print)

Journal current issue

Vol. 31, Iss. 1
January 2025

Archive

Видите ошибку в отчестве? Отключите перевод, это английская версия сайта!