基于指数平滑法和马尔科夫模型的零售额预测

摘要

Based on The actual value,commodity retail sales of retail business of Guangdong Province in 2012 were predicted by using exponential smoothing method.Taking -25%、-20、-15%、-10%as division threshold values,the relative errors of prediction results by using exponential smoothing method were divided into four state intervals,the prediction results of exponential smoothing method were modified by using Markov model,and compared with exponential smoothing method.The proposed the prediction results of exponential smoothing method were modified by using Markov model.The proposed method has higher precision to meet the actual demands.%该文基于零售业商品零售额的实际值，应用指数平滑法进行预测了2012年零售业零售额，以-25％、-20、-15％、-10％为划分阈值，将指数平滑法预测结果的相对误差划分为4个状态区间，应用马尔科夫模型对指数平滑法的预测结果进行修正，并与指数平滑法的预测结果进行比较。最后用马尔科夫模型进行修正并对预测结果加以评价。提出的方法精度较高，满足实际需求。