APPLICATION OF VECTOR AUTOREGRESSIVE MODELS FOR THE ANALYSIS OF DYNAMICS AND FORECASTING OF MACROECONOMIC PROCESSES IN UKRAINE

Abstract

The article provides a comprehensive analysis and evaluation of the effectiveness of using vector autoregressive models for modeling and forecasting macroeconomic processes in Ukraine. It is substantiated that under conditions of a high level of uncertainty and structural transformations, traditional econometric approaches often have a limited ability to account for complex interrelationships, which reduces the accuracy of forecasts. Instead, the use of vector autoregressive models allows for the investigation of dynamic interdependencies between variables, treating them as a single system of endogenous indicators.

Based on official statistical data, an information base for the research was formed, including indicators of gross domestic product, consumer price index, lending rate, and real effective exchange rate. Using the construction of a Pearson correlation matrix, key empirical patterns were identified, in particular, a critically high direct dependence between gross domestic product, consumer spending, and imports, indicating a structural vulnerability of the economy and the problem of multicollinearity. A powerful pass-through effect of the exchange rate to domestic inflation and stagflationary markers, where price increases suppress industrial production, was also identified. Additionally, a graphical analysis of the indicators' development trajectories was conducted, which allowed for the visualization of the synchronicity of their movement and the identification of significant structural breaks caused by the crisis shocks of 2014–2015 and 2022.

The research methodology involved the transition to stationary time series through log-differencing, the stability of which was confirmed by the augmented Dickey-Fuller test. The quality of the constructed vector autoregressive model was verified using the backtesting method on test data for the years 2024–2025. The calculated mean absolute percentage error, which amounted to 1.29% for the consumer price index and 4.4% for gross domestic product, confirms the high accuracy and feasibility of using the chosen tools. The generated forecast for 2026–2028 indicates a gradual slowdown of inflation to 5% and stable growth of the nominal gross domestic product.


References

1. Coulombe P. G., Leroux M., Stevanovic D., Surprenant S. (2022). How is Machine Learning Useful for Macroeconomic Forecasting? Journal of Applied Econometrics, Vol. 37, No. 5. https://doi.org/10.1002/jae.2910
2. Shcherbinina, S. O., & Shevchenko, O. M. (2025). Artificial intelligence in economic-mathematical modeling of entrepreneurship development scenarios in Ukraine. Ekonomichnyi Prostir (Economic Space), No. 201. https://doi.org/10.30838/EP.201.241-245 [in Ukrainian].
3. Blanchard O. (2018). On the Future of Macroeconomic Models. Oxford Review of Economic Policy, Vol. 34, No. 1/2. https://www.jstor.org/stable/48539406
4. Sengupta S., Chakraborty T., Singh S. K. (2025). Forecasting CPI Inflation Under Economic Policy and Geopolitical Uncertainties. International Journal of Forecasting, Vol. 41, No. 3. https://doi.org/10.1016/j.ijforecast.2024.08.005
5. Otto P., Doğan O., Taşpınar S. (2024). A Dynamic Spatiotemporal Stochastic Volatility Model with an Application to Environmental Risks. Econometrics and Statistics, Vol. 30. https://doi.org/10.1016/j.ecosta.2023.11.002
6. Sims C. A. (1980). Macroeconomics and Reality. Econometrica, Vol. 48, No. 1. https://doi.org/10.2307/1912017
7. Lütkepohl H. (2005). New Introduction to Multiple Time Series Analysis. Springer-Verlag. https://doi.org/10.1007/978-3-540-27752-1 (
8. Giannone D., Lenza M., Primiceri G. E. (2015). Prior Selection for Vector Autoregressions. Review of Economics and Statistics, Vol. 97, No. 2. https://doi.org/10.1162/REST_a_00483
9. Wołoszyn J., Bukowski S. (2025). The Impact of AI on Economic Modelling. European Research Studies Journal, Vol. 28, No. 1. https://doi.org/10.35808/ersj/3927
10. Ameziane K., Benyacoub B. (2022). Exchange Rate Volatility Effect on Economic Growth Under Different Exchange Rate Regimes: New Evidence from Emerging Countries Using Panel CS-ARDL Model. Journal of Risk and Financial Management, Vol. 15, No. 11. https://doi.org/10.3390/jrfm15110499
11. Grui A., Lysenko R. (2017). Nowcasting Ukraine's GDP Using a Factor-Augmented VAR (FAVAR) Model. Visnyk of the National Bank of Ukraine, No. 242. https://doi.org/10.26531/vnbu2017.242.005
12. Takahashi Y., Otaka K., Kato N. (2025). Potential Applications of Generative AI in Economic Simulations. Bank of Japan Research LAB, Vol. 25, No. 1. https://www.boj.or.jp/en/research/wps_rev/lab/lab25e01.htm
13. Kole E., Dijk D. (2023). Moments, Shocks and Spillovers in Markov-Switching VAR Models. Tinbergen Institute Discussion Paper, No. TI 2021-080/III. https://doi.org/10.2139/ssrn.3924951
14. Lukianenko I., Nasachenko M., Tokarchuk T. (2025). Estimating Inflation and Inflation Expectations Based on a Markov-Switching Vector Autoregression Approach: Case of Ukraine. Financial and Credit Activity: Problems of Theory and Practice, Vol. 3, No. 62. https://doi.org/10.55643/fcaptp.3.62.2025.4724
15. Shapovalenko N. (2021). A BVAR Model for Forecasting Ukrainian Inflation and GDP. Visnyk of the National Bank of Ukraine, No. 251. https://doi.org/10.26531/vnbu2021.251.02
16. Kramar V., Chepyha B. (2025). Anchoring Firms’ Inflation Expectations in Ukraine: Assessing Shock and Level Anchoring with a VAR Approach. Visnyk of the National Bank of Ukraine (Occasional Paper), No. 2025/06. https://doi.org/10.26531/vnbu2025.op06
17. Aastveit K. A., Bjørnland H. C., Cross J. L. (2024). Global VAR Models for Open Economies: A New Perspective on Trade and Inflation. Journal of International Economics, Vol. 147. https://doi.org/10.1016/j.jinteco.2023.103856
18. Faryna O., Simola H. (2019). How Trade Composition Affects Sensitivity to Foreign Shocks: Applying a Global VAR Model to Ukraine. Visnyk of the National Bank of Ukraine, No. 247. https://doi.org/10.26531/vnbu2019.247.01
19. Grui A., Vdovychenko A., Gurnytska S., Myronova L. (2023). Anchoring of Inflation Expectations in Ukraine: Evidence from Firm-Level and Macro Data. National Bank of Ukraine Working Paper, No. 2023/02. https://doi.org/10.26531/nbubwp2023-02
20. Vavra J. (2025). Non-linear Macroeconomic Dynamics and Regime Switching in Emerging Markets. Springer Economics, Vol. 12, No. 1. https://doi.org/10.1007/s40503-024-00123-x
21. Abdel Wahed M. (2026). Hybrid Econometric and Machine Learning Approaches for Robust Anomaly Detection. In M. Rahmouni (Ed.), Robust Methods for Anomaly Detection in Econometrics. IGI Global Scientific Publishing. https://doi.org/10.4018/979-8-3373-8297-5.ch007
22. Kovbatiuk M., & Semeniuk D. (2025). Comparative Analysis of Trend Methods for Forecasting Economic Indicators on the Example of the Economy of Ukraine. Tavrian Scientific Bulletin. Series: Economics, No. 24, pp. 111–120. https://doi.org/10.32782/2708-0366/2025.24.11 [In Ukrainian].
23. World Bank Open Data : website. URL: https://data.worldbank.org
24. International Financial Statistics : website / IMF. URL: https://data.imf.org
25. Official website of the National Bank of Ukraine : website. URL: https://bank.gov.ua
26. State Statistics Service of Ukraine : website. URL: http://www.ukrstat.gov.ua