Application of cigarette sales forecasting model based on time series data
应用科学进展 · 15.04.2026 · 3 (2)
DOI: 10.69928/2026000201 ↗Kurz gefasst
Fachartikel in der Zeitschrift 应用科学进展, online erschienen am 15.04.2026. Verfasst von Luyu Jin und 1 weiteren Autor:innen. Die Arbeit nennt 11 Literaturangaben.
Abstract (Originalsprache)
With the transformation of the cigarette marketing model, the cigarette sales market has gradually changed from "seller's market" to "buyer's market". Pre-intervention is far more important than post-adjustment. Market expectations are obtained through data analysis, and operation regulation is carried out in combination with various objectives and tasks. It is particularly important to regulate the rhythm of cigarette sales, maintain the stable and orderly economic operation, and optimize the state of the regional market. This paper selects the monthly sales data of different categories in two regions, uses the grey forecasting model, ARIMA model and BP neural network model to forecast the historical sales data, and tests the forecast results. The prediction results show that the prediction error of BP neural network is relatively small. The prediction errors of the grey prediction model and the ARIMA model are close to each other, in which the grey prediction model is more suitable for the prediction of the data with large changes in similar months, and the ARIMA model is more suitable for the time series data with relatively stable changes. Moreover, the three prediction results all show that the prediction errors have spatiotemporal heterogeneity. At the same time, the sales forecasting error of different categories of cigarettes is also slightly different, and the three forecasting results generally show that the sales error of Class B cigarettes is larger than that of class A cigarettes. On the whole, BP neural network model is more suitable for cigarette sales forecasting. At the same time, considering the forecasting error of historical data, the average error and standard deviation can be calculated to determine the sales range and its probability, which will be a good guide for cigarette delivery, reduce social inventory, promote the stable development of economic operation and maintain a good market state.
Zitieren
APA
Jin, L., & Xue, F. (2026). Application of cigarette sales forecasting model based on time series data. 应用科学进展, 3(2). https://doi.org/10.69928/2026000201
Vancouver
Jin L, Xue F. Application of cigarette sales forecasting model based on time series data. 应用科学进展. 2026;3(2). doi:10.69928/2026000201
BibTeX
@article{Jin2026,
author = {Jin, Luyu and Xue, Fangfang},
title = {Application of cigarette sales forecasting model based on time series data},
journal = {应用科学进展},
year = {2026},
volume = {3},
number = {2},
doi = {10.69928/2026000201}
}
EUVIMED