FORECASTING THE ECONOMIC INDICATORS OF A TRADING ENTERPRISE TAKING INTO ACCOUNT THE SEASONALITY OF SALES
DOI:
https://doi.org/10.32782/IT/2024-1-11Keywords:
forecasting, trend, seasonality, Fourier series, model, trading enterprise.Abstract
The forecast of sales volume and other economic indicators are important components when planning the activities of trading organizations. It is based on several factors: historical data, industry and market trends, the current state of the sales funnel. The accuracy of the forecast depends on two most important elements: the collection of reliable initial data and obtaining the correct conclusions from them. Having overestimated the dynamics of growth, the company may needlessly invest funds in increasing stocks or team size. An underestimation of the sales volume can lead to a deficit and, as a result, a partial loss of potential profit. When building a sales forecast, the assessment of seasonality of demand plays an important role. Accounting for seasonality complicates the task of forecasting by increasing the number of variables, which makes it difficult to use complex models in the practice of business management. Therefore, the purpose of this work is to develop a practical algorithm for building a forecast of economic indicators of a trading company with pronounced seasonality of sales, based on the use of various forecasting approaches using Fourier series. The developed algorithm uses the values of some economic indicator of a trading company for several previous years. This period is divided into two parts: a set of calibration data and a set of verification data, which will allow to analyze the quality of the results of forecast construction. Data from the calibration set are approximated by Fourier series (for each year separately). A series is built, the coefficients of which are exponentially smoothed values of the coefficients of the Fourier series, taken from the calibration set. Next, a forecast is formed for the verification period, during the construction of which the coefficients of the Fourier series are selected taking into account the minimization of the error of the deviations of the forecast values of the economic indicator from the values from the verification period. The scientific novelty of this work consists in the development of a methodology for forecasting economic indicators, based on the approximation of time series by Fourier series for several previous years, finding the average smoothed value of the coefficients of Fourier series and building an effective forecast of the trend-seasonal model of indicators of a trading company. The created algorithm was tested on real data of a wholesale trade enterprise and implemented in the practice of procurement planning. The forecast built with the help of this algorithm is planned to be used in the future to create a management system for the company's balances using discrete control systems with state vector observers.
References
Fildes R, Goodwin P, Lawrence M, Nikolopoulos K. Effective forecasting and judgmental adjustments: an empirical evaluation and strategies for improvement in supply-chain planning. International Journal of Forecasting. 2009. Vol. 25(1). P. 3–23.
Acara Y, Gardner ES. Forecasting method selection in a global supply chain. International Journal of Forecasting. 2012. Vol. 28(4). P. 842–848.
Xia M, Zhang Y, L. W, Ye X. Fashion retailing forecasting based on extreme learning machine with adaptive metrics of inputs. Knowledge-Based Systems. 2012. Vol. 36. P. 253–259.
Smith C.D, Mentzer J.T. Forecasting task technology fit: The influence of individuals, systems and procedures on forecast performance. International Journal of Forecasting. 2010. Vol. 26(1). P. 144–161.
Holt CC. Forecasting seasonal and trends by exponentially weighted moving averages. International Journal of Forecasting. 2004. Vol. 20(1). P. 5–10.
Ймовірнісно-статистичні методи моделювання і прогнозування: наук.-навч. вид. / Гуськова В. Г., Бідюк П. І., Гасанов А. С. Київ: Видавництво НПУ імені М. П. Драгоманова, 2022. 456 с.
W. Press; S. Teukolsky, W. Vetterling, B. Flannery. Chapter 12. Fast Fourier Transform. Numerical Recipes: The Art of Scientific Computing. New York, USA:Cambridge University Press. 2007. ISBN 978-0-521-88068-8.
A. Fumi, A. Pepe, L. Scarabotti and Massimiliano M. Schiraldi. Fourier Analysis for Demand Forecasting in a Fashion Company. International Journal of Engineering Business Management. 2013. Vol. 5(37). DOI:10.5772/56840
Li M., Liu Y., Zhi S., Wang T., Chu F. Short-time Fourier Transform Using Odd Symmetric Window Function. Journal of Dynamics, Monitoring and Diagnostics, 2021. Vol. 1(1). P. 37–45. DOI: 10.37965 https://doi.org/10.37965/jdmd.v2i2.39
L. Ye, N. Xie, J. E. Boylan, Z. Shang, Forecasting seasonal demand for retail: A Fourier time-varying grey model. International Journal of Forecasting. 2024. DOI: https://doi.org/10.1016/j.ijforecast.2023.12.006.
M. Dekker, K. van Donselaar, P. Ouwehand. How to use aggregation and combined forecasting to improve seasonal demand forecasts, International Journal of Production Economics, Vol. 90, Issue 2, 2004, P. 151–167, https://doi.org/10.1016/j.ijpe.2004.02.004.
Ord K., Fildes R. A., Kourentzes N. Principles of business forecasting: 2nd ed. Wessex Press Publishing Co. 2017. 588 p.
Brown R. G. Smoothing forecasting and prediction of discrete time series. New York. Dover Publications. Dover Phoenix Ed. 2004.480 p.
