摘要
准确的销售预测可以帮助企业制定正确的营销策略以减小企业在经营过程的损失。由于快销服装行业的特殊性,产品生命周期通常较短且销售受诸多非线性因素影响,传统的销售预测模型精准度往往差强人意。论文基于Prophet算法的基础上,优化时间序列分解的各因式项,结合电商平台历年的销售数据,依据产品生命周期特性构建模型,结果表明,改进后的Prophet算法模型预测准确性高于传统模型,且耗时有所降低。
Accurate sales forecasts can help companies develop the right marketing strategy to reduce the loss of business in the business process.Due to the particularity of the fast-selling apparel industry,the product life cycle is usually short and sales are affected by many non-linear factors.The accuracy of traditional sales forecasting models is often unsatisfactory.Based on the Proph⁃et algorithm,this paper optimizes the various factorial items of time series decomposition,combines the sales data of E-commerce platform over the years,and builds a model based on product life cycle characteristics.The results show that the improved Prophet algorithm model has higher prediction accuracy than traditional ones,and time-consuming is reduced.
作者
苏新
SU Xin(Jiangsu University of Science and Technology,Zhenjiang 212000)
出处
《计算机与数字工程》
2021年第6期1258-1261,共4页
Computer & Digital Engineering