基于BP神经网络模型的商洛市卷烟需求预测

摘要

Under the background of the new normal economy,the tobacco industry's own "microclimate" inevitably changes,and the economic operation of the industry presents some new changes and characteristics.According to the relevant data collection and investigation of cigarette consumption market in Shangluo City,established cigarette demand neural network forecasting model,introduced the index selection,clustering and screening,used Matlab software for data simulation and training.Compared with the actual sales,it is proved that the BP neural network prediction results are accurate.%在国家经济新常态的大背景下,烟草行业自身“小气候”不可避免地发生变化,行业经济运行呈现出一些新的变化和特点.根据商洛市卷烟消费市场调查收集的有关数据,建立卷烟需求神经网络预测模型,介绍了其指标选取、聚类和筛选,利用Matlab软件对数据进行训练、仿真.与实际销量进行对比分析,证明采用的BP神经网络预测结果准确.