基于灰色-马尔可夫模型的国内游客总数预测  

Prediction of Total Domestic Tourists Based on Grey-Markov Model

作　　者：张冬咏[1] 陈泗达 ZHANG Dongyong;CHEN Sida(College of Information and Management Science,Henan Agricultural University,Zhengzhou 450002,China)

机构地区：[1]河南农业大学信息与管理科学学院,郑州450002

出　　处：《河南科学》2020年第1期96-101,共6页Henan Science

基　　金：河南省教育厅软科学项目(18B630008);河南科技智库(HNKJZK-2019-48B)。

摘　　要：结合灰色模型在小样本下预测精度较高的优点和马尔可夫模型对随机波动数据处理结果较好的优点,以2004—2015年国内游客总数构建传统灰色GM(1,1)模型、无偏灰色GM(1,1)模型、灰色马尔可夫模型和无偏灰色马尔可夫模型,并对比2016—2018年国内游客总数的预测值与实际值.结果表明,灰色马尔可夫模型和无偏灰色马尔可夫模型相比于传统灰色GM(1,1)模型和无偏灰色GM(1,1)模型的平均相对误差分别提高了2.36个百分点和2.33个百分点,灰色模型结合马尔可夫模型后能够解决对随机波动数据的预测偏差,有效提高预测精度.Combined with the advantage of the grey model with higher prediction accuracy under small sample and the advantage of Markov model for better processing resutts of random fluituation data,the traditional grey GM(1,1)model,unbiased grey GM(1,1)model,grey-Markov model and unbiased grey-Markov model are constructed with the total number of domestic tourists from 2004 to 2015.The total number of domestic tourists from 2016 to 2018 is selected to compare and analyze its prediction accuracy.The results show that the average relative error of grey-Markov model and unbiased grey-Markov model is 2.36%and 2.33%higher than that of traditional grey GM(1,1)model and unbiased grey GM(1,1)model,respectively.And the grey model combined with Markov model can solve the prediction deviation of random fluctuation data and effectively improve the prediction accuracy.

关 键 词：GM(1 1)模型 马尔可夫模型 预测 

分 类 号：F291.1[经济管理—国民经济]