Effectiveness of Using a Constant Coefficient in Combining Forecasts to Improve Forecasting Accuracy
https://doi.org/10.26794/2408-9303-2024-11-4-96-107
Abstract
The article discusses the methodology for using a constant coefficient when combining forecasts. Today, there are many options for constructing weighting coefficients, and some of them include a constant coefficient in the combination due to the assumption that it improves forecasting accuracy. Unfortunately, there is no clear and unambiguous answer to the question of how true this hypothesis is — it has both positive and negative sides. The purpose of the study is to determine the advantages and disadvantages of using a constant coefficient when combining forecasts based on available practical and theoretical data, as well as to form a unified approach to this issue. In the course of the work, scientific methods for combining forecasts were applied (proposed by K. Granger and R. Ramanathan), one of which involves the presence and calculation of the constant coefficient. The practical results obtained by the author of the article have generally confirmed the value of including a constant coefficient in the combined forecast, on the basis of which it was concluded that the use of the latter is possible if there is confidence that it can improve the accuracy of forecasting. The study also identified the need to find such an approach to constructing weighting coefficients that would take into account the possibility of changing the constant coefficient for combining forecasts, thereby expanding the possibilities of its application.
About the Author
A. A. Surkov
Institute of Economics RAS
Russian Federation
Russian Federation
Anton A. Surkov — Cand. Sci. (Econ.), Senior Researcher at the Center for Macroeconomic Analysis and Forecasting
Moscow
References
1. Eskindarov M. A., Salin V. N., Melnik M. V., Mikhnenko O. E. Convergence of economic and statistical accounting and analysis in the social sphere. Uchet. Analiz. Audit. = Accounting. Analysis. Auditing. 2024;11(2):6–23. (In Russ.). DOI: 10.26794/2408–9303–2024–11–2–6–23
2. Frenkel A. A., Surkov A. A. Determination of weighting coefficients when combining forecasts. Voprosy statistiki = Statistics questions. 2017;12:3–15. (In Russ.).
3. Bates J. M., Granger C. W.J. The combination of forecasts. Operational Research Quarterly. 1964;4(20):451–468. DOI: 10.2307/2982011
4. Newbold P., Granger C. W.J. Experience with forecasting univariate time series and the combination of forecasts. Journal of the royal statistical society. 1974;137:131–164. DOI: 10.2307/2344546
5. Makridakis, S., Spiliotis E., Assimakopoulos V. The M4 competition: Results, findings, conclusion and way forward. International Journal of Forecasting. 2018;4(34):802–808. DOI: 10.1016/j.ijforecast.2018.06.00
6. Frenkel A. A. Mathematical methods for analyzing the dynamics and forecasting labor productivity. Moscow: Ekonomika; 1972. 190 p. (In Russ.).
7. Ershov E. B. About one method for combining partial forecasts. Uchenye zapiski po statistike = Scientific notes on statistics. 1973; XXII–XXIII:87–105. (In Russ.).
8. Styrin K. Forecasting inflation in Russia using the method of dynamic averaging of models. Dengi i kredit = Money and credit. 2019;1:3–18. (In Russ.). DOI: 10.31477/rjmf.201901.03
9. Surkov A. A. Combining economic forecasts using expert information. Statistika i ekonomika = Statistics and economics. 2019;5:4–14. (In Russ.). DOI: 10.21686/2500–3925–2019–5–4–14
10. Frenkel A. A., Surkov A. A. Forecast fusion is an effective tool for improving forecasting accuracy. Moscow: URSS; 2023. 200 p. (In Russ.).
11. Lazcano A., Herrera P. J., Monge M.A Combined model based on recurrent Neural Networks and graph convolutional networks for financial time series forecasting. Mathematics. 2023;1(11). DOI: 10.3390/math11010224
12. Armstrong J. S. Combining forecasts. International Series in Operations Research & Management Science 2001;30:417–439. DOI: 10.1007/978–0–306–47630–3
13. Surkov A. A. Are negative weights bad for combining forecasts? Statistika i ekonomika = Statistics and economics. 2023;4:4–11. (In Russ.). DOI: 10.21686/2500–3925–2023–4–4–11
14. Granger C. W.J., Ramanathan R. Improved methods of combining forecasts. Journal of Forecasting. 1984;3:197–204. DOI: 10.1002/for.3980030207
15. Clemen R. T. Linear constraints and the efficiency of combined forecasts. Journal of Forecasting. 1986;1(5):31–38. DOI: 10.1002/for.3980050104
16. Trenkler G., Liski E. P. Linear constraints and the efficiency of combined forecasts. Journal of Forecasting. 1986;3(5):197–202. DOI: 10.1002/for.3980050306
17. Hsiao C., Wan S. K. Is there an optimal forecast combination? Journal of econometrics. 2014;2(178):294–309. DOI: 10.1016/j.jeconom.2013.11.003
18. Terui N., Dijk H. K. Combined forecasts from linear and nonlinear time series models. International Journal of Forecasting. 2002;3(18):421–438. https://doi.org/10.1016/S0169–2070(01)00120–0
19. Holmen J. S. A note on the value of combining short-term earnings forecasts: A test of Granger and Ramanathan. International Journal of Forecasting. 1987;2(35):239–243.
20. Celal A., Sevket I. G. An empirical analysis of the accuracy of SA, OLS, ERLS and NRLS combination forecasts. International Journal of Forecasting. 1991;1(8):27–43.
21. Gökhan S. The Combination of forecasts: An application of a time-varying simple weighting method to inflation forecasts in Turkey. International Research Journal of Applied Finance. 2022;3(2):270–301
Review
For citations:
Surkov A.A. Effectiveness of Using a Constant Coefficient in Combining Forecasts to Improve Forecasting Accuracy. Accounting. Analysis. Auditing. 2024;11(4):96-107. (In Russ.) https://doi.org/10.26794/2408-9303-2024-11-4-96-107
Views:
113
.png)
.png)
Email this article 
