Keywords
social needs forecasting
crisis conditions
refugees
international economic relations
quantitative models
expert evaluations
social support
How to Cite
Abstract
In the modern world, global crises such as wars, pandemics, and migration flows pose significant challenges to socio-economic planning. Specifically, unstable conditions limit the effectiveness of traditional methods for forecasting social needs, which rely on historical data. This article examines the application of a hybrid forecasting method that combines quantitative time-series models with expert evaluations. The study aims to enhance the accuracy of forecasting and the adaptability of social programs under crisis conditions. The developed hybrid method consists of two main components: a forecasting block, utilizing quantitative time-series models (ARIMA, SARIMA), and an expert block, integrating qualitative assessments from specialists. The application of this method allows for the consideration of sudden changes in socio-economic conditions, which are not accounted for by traditional models. International experience demonstrates the success of the hybrid approach. For instance, in Syria, it facilitated the timely planning of humanitarian aid, particularly in healthcare and food supply. In Poland and Germany, the method has been used to forecast refugees' needs for housing, education, and labor market integration. The study reveals that the hybrid method significantly reduces forecasting errors compared to traditional approaches. A model example of estimating housing service needs for internally displaced persons under conflict conditions shows a reduction in average forecast error from 2.33% (ARIMA method) to 1.97% through the use of the hybrid approach. This ensures more accurate planning of social resources, improving the effectiveness of support programs. The hybrid forecasting method is an effective tool for determining social needs in crisis conditions. Its application ensures forecast accuracy, model flexibility, and adaptability to emerging conditions. The study results confirm the feasibility of employing this method in international social support programs, particularly for countries experiencing armed conflicts or hosting significant numbers of refugees. It is recommended to integrate hybrid approaches into the practices of international organizations such as the United Nations and the European Union, as well as into national socio-economic planning strategies. Doing so will reduce social tension, optimize resource allocation, and contribute to achieving sustainable development.
References
1. Dobrovolska, O., Kolotilina, O., & Ostapenko, M. (2024). Forecasting macroeconomic dynamics in Ukraine: The impact of a full-scale war. SocioEconomic Challenges (SEC), 8(3), 211–237. https://doi.org/10.61093/sec.8(3).211-237.2024
2. Wicke, L., Dhami, M. K., Önkal, D., & Belton, I. K. (2022). Using scenarios to forecast outcomes of a refugee crisis. International Journal of Forecasting, 38(3), 1175–1184. https://doi.org/10.1016/j.ijforecast.2019.05.017
3. Durairaj, D. M., & Mohan, B. K. (2022). A convolutional neural network-based approach to financial time series prediction. Neural Computing and Applications, 34(16), 13319–13337. https://doi.org/10.1007/s00521-022-07143-2
4. Mulesa, O., Geche, F., Batyuk, A., & Buchok, V. (2017, September). Development of combined information technology for time series prediction. In Conference on Computer Science and Information Technologies (pp. 361–373). Cham: Springer International Publishing. https://doi.org/10.1007/978-3-319-70581-1_26
5. Abdollahi, H. (2020). A novel hybrid model for forecasting crude oil price based on time series decomposition. Applied Energy, (267), 115035. https://doi.org/10.1016/j.apenergy.2020.115035
6. Willekens, F. (1994). Monitoring international migration flows in Europe: Towards a statistical database combining data from different sources. European Journal of Population/Revue Européenne de Démographie, 10(1), 1–42. https://doi.org/10.1007/BF01268210
7. Chukwuere, J. E. (2024). Conceptualizing predictive conceptual model for unemployment rates in the implementation of Industry 4.0: Exploring machine learning techniques. arXiv preprint. https://doi.org/10.32388/E0D5TO.2
8. Thayyib, P. V., Thorakkattle, M. N., Usmani, F., Yahya, A. T., & Farhan, N. H. (2023). Forecasting Indian goods and services tax revenue using TBATS, ETS, neural networks, and hybrid time series models. Cogent Economics & Finance, 11(2), 2285649. https://doi.org/10.1080/23322039.2023.2285649
9. Kykyna, Y., & Mulesa, O. (2024). Hybrid method of forecasting social service needs. Science and Technology Today, 5(33), 1190–1199. https://doi.org/10.52058/2786-6025-2024-5(33)-1190-1199
10. Bennett, C. (2002). United Nations Office for the Coordination of Humanitarian Affairs (UNOCHA) orientation handbook. United Nations Office for the Coordination of Humanitarian Affairs (UNOCHA). https://digitallibrary.un.org/record/503265?v=pdf
11. Ahmad, A., Rassa, N., Orcutt, M., Blanchet, K., & Haqmal, M. (2021). Urgent health and humanitarian needs of the Afghan population under the Taliban. The Lancet, 398(10303), 822–825. https://doi.org/10.1016/S0140-6736(21)01963-2
12. Zakirova, S. (2023). Ukrainian refugees from the war: Analysis of informational support resources. Academic Papers of The Vernadsky National Library of Ukraine, (67), 54–81. https://doi.org/10.15407/np.67.054 (in Ukrainian)
13. Horobets, O. (2015). Ukrainian migrants in Poland: Social aspects of resource use. In Proceedings of the Fourth International Scientific and Practical Conference in Memory of Honorary Professor of TNTU, Academician of the NAS of Ukraine Chumachenko Mykola Grygorovych "Innovative Aspects of Resource Use" (pp. 133–134) (in Ukrainian)
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Copyright (c) 2024 Yevhenii Kykyna