神经网络在卷烟销售量预测的应用研究

Application of Cigarette Sales Forecasting Based on Neural Network

研究卷烟销售预测准确问题,卷烟销售量具有季节性和周期性动态变化规律,并受经济、人口等因素的影响,使系统存在明显的非线性特征,波动范围比较大,传统线性预测模型难以准确预测。为了提高卷销售预测精度,提出一种能够反映卷烟销售量变化规律的Elman神经网络的卷烟销售预测模型。首先采用逐步拓阶方法确定卷烟销售量的最佳滞后阶数,然后利用最佳滞后阶数最对卷烟销售数据进行重组,并输入Elman神经网络学习,利用Elman神经网络的动态和反馈特点对卷烟销售量进行预测。将建立的模型应用于云南某烟草公司某种卷烟销售的预测,结果表明,Elman神经网络模型有效提高了卷烟销售预测精度,降低了预测误差,为烟草行业销售管理预测提供科学依据。

Research about cigarette sales forecast problems.Cigarette sales are influeced by seasonal and periodic fluctuation,therefore,the traditional models are difficult to describe the dynamic change regulation accurately,and the forecast precision is low.In order to improve the forecasting accuracy,a cigarette sales forecast model was proposed based on Elman neural network.Firstly,the best cigarette sales lagging order was obtained.Then restructure data were input into Elman neural network for training.Lastly,The dynamic and feedback characteristics of Elman neural networkwere used to forecast the cigarette sales.The model was tested by a Yunnan tobacco company＇s cigarette sales data.The results show that,compared with other forecasting model,Elman neural network model can improve the cigarette sales forecast precision,and reduce the forecasting error.The forecasting results can be used for tobacco industry management.