Maryna Abramova, PhD (Economics), Senior Researcher, Senior Researcher at the Central Research Institute of the Armed Forces of Ukraine, Kyiv, Ukraine

ORCID ID: 0000-0001-7644-9988
e-mail: Elaira3@gmail.com

Iryna Chernyshova, Candidate of Military Sciences, Senior Researcher, Senior Researcher at the Central Research Institute of the Armed Forces of Ukraine, Kyiv, Ukraine

ORCID ID: 0000-0002-5958-7059
e-mail: i_ttv@ukr.net

Serhii Zhurenko, student at the Institute of Logistics and Support of Troops of the National Defense University of Ukraine, Kyiv, Ukraine

ORCID ID: 0009-0003-1598-935X
e-mail: SZhurenko_1978@gmail.com

Dmytro Vislenko, student at the Institute of Logistics and Support of Troops of the National Defense University of Ukraine, Kyiv, Ukraine

ORCID ID: 0009-0007-0320-6732
e-mail: vislenkodv_1985@gmail.com

The Importance of Taking into Account the Results of Additive Modeling During Dynamics Process Analyzing

Abstract. Introduction. In order to cope with such changes during the analysis, it is necessary to understand the essence of the methods and approaches that should be used by the researcher. The authors of this article aim to inform a wide range of scientists about the possibilities of using additive modeling not only to forecast data with seasonal and random components, but also to take into account its results at each stage of analysis to determine the essence (features) of the dynamics of processes (on the example of the gross domestic product of the Russian Federation and Ukraine until 2026), as well as to compare its forecast data with the results of extrapolation using a sixth-order polynomial. The authors of this paper extrapolated the data of the gross domestic product of the Russian Federation and Ukraine to 2026 using additive modeling.

Purpose. The purpose of the article is to convey to a wide range of scientists the possibilities of using additive modeling not only for forecasting data taking into account the seasonal component, but also taking into account its results at each stage of analysis to determine the essence (features) of the dynamics of processes (on the example of data on the gross domestic product of the Russian Federation and Ukraine with a forecast until 2026), as well as a comparison of its forecast data with the results of extrapolation using a sixth-order polynomial.

Results. Thus, based on the results of calculating the average percentage error, average absolute percentage error, coefficient of determination and the results of checking the adequacy of the model, the following conclusions were made: the specificity of the statistical data proposed for analysis (GDP of Ukraine and the Russian Federation) fully meets the requirements of additive modeling; there are no significant deviations from the average variation in the statistical data proposed for analysis, so the reliability of extrapolation is high; similar values in the changes in the average percentage error, average absolute.

Conclusions. That is, to bring the extrapolation closer to reality, the value of the seasonal component was calculated. According to the results obtained, the dynamics of Ukraine’s GDP, based on the data for 1986-2022, is more stable than that of the Russian Federation (1987-2022). According to the results obtained after evaluating the results of the two calculations, the following conclusions can be drawn: although Ukraine’s GDP indicators reacted more sharply to changes in the national economy (sharper drops in the sample than in the Russian Federation’s indicators), the extrapolation data of both approaches have a similar trend, which is evidence of a faster recovery of economic processes within the country; the results of extrapolation of the two approaches do not coincide, which indicates the presence of hidden processes in the national economy that have a significant impact on the dynamics of Thus, today the problem is not only in more realistic data forecasting, but also in the qualitative interpretation of the results at each stage of analysis to determine the essence (features) of the dynamics of processes.

Keywords: additive modeling; extrapolation; gross domestic product; forecasting.

References:

1. Friedman, J. & Stuetzle, W. (1981). Projection Pursuit Regression. Journal of the American Statistical Association, 76, 817-823 [in English].
2. Roshko, N. B. (2013). Forecasting the income of tourist enterprises based on additive modeling. Vcheni zapysky universytetu KROK. Seriya: Ekonomika, 34, 292-300 [in Ukrainian].
3. Mazurenko, V. P. & Kondratchuk, K. S. (2011). Modeling the dynamics of mergers and acquisitions in the world economy (on the example of European countries). Actual Problems of International Relations, 1, 100-120 [in Ukrainian].
4. Dobulyak, L. P. & Kostenko, S. B. (2019). Vykorystannya trendovykh modeley dlya doslidzhennya tendentsiy rozvytku maloho pidpryyemnytstva v Ukrayini. Uzhhorod : Helvetyka [in Ukrainian].
5. Bidyuk, P. I. (2014). Ymovirnisno-statystychni metody modelyuvannya i prohnozuvannya. Mykolayiv, ChDY Petra Mohyly [in Ukrainian].
6. Lukashyn, Y. P. (2003). Adaptyvnye metody kratkosrochnoho prohnozyrovanyya vremennykh ryadov. Fynansy i statystyka [in Ukrainian].
7. Dorozhenko, L. I. (2014). The essence of cost optimization using economic and mathematical methods. Naukovyy visnyk Khersonsʹkoho derzhavnoho universytetu. Seriya: «Ekonomichni nauky», 228-231 [in Ukrainian].
8. Chumachenko, D. I. & Chumachenko, T.O. (2020). Mathematical models and methods of epidemic processes forecasting [in English].
9. Hlushkov, O. V. (2015). Development and application of cybernetic methods to study the dynamics of hierarchical chaotic processes in quantum, informative and geophysical systems. Оdessa. Retrieved from http://eprints.library.odeku.edu.ua/id/eprint/2149/1/Zvit_K_Glushkov_0111U005226_2015.PDF [in Ukrainian].
10. Revutsʹka, L. O. & Bidyuk, P. I. (2020). Mathematical modeling and forecasting of non-linear non-stationary processes in economics and finance. Problemy informatyzatsiyi: materialy VIII mizhnarodnoyi nauk.-tekhn. konf. Cherkasy, 21 [in Ukrainian].
11. Polokhova, I. M. (2014). Macroeconomic forecasting of Ukraine’s real GDP indicators. Visnyk Kyyivsʹkoho natsionalʹnoho universytetu tekhnolohiy ta dyzaynu, 4 (78), 18-25 [in Ukrainian].
12. Pestovsʹka, Z. S. (2011). Methods of forecasting economic indicators: macro and micro levels. Yevropeysʹkyy vektor ekonomichnoho rozvytku, 5, 151-157 [in Ukrainian].
13. Hastie, T. J. & Tibshirani, R. J. (1990). Generalized Additive Models. New York: Chapman & Hall [in English].
14. Stone, C. J. (1985). Additive Regression and Other Nonparametric Models. Annals of Statistics, 13, 689–705 [in English].
15. Palahin, V. V. & Ivchenko, O. V. (2009). Adaptation of the polynomial maximization method for estimating the parameters of random variables using a statistically dependent sample. Systemy obrobky informatsiyi, 2, 118-123 [in Ukrainian].
16. World Development Indicators (2022). Retrieved from https://databank.worldbank.org/reports.aspx?source=2&country [in English].