基于GM-ES-GASVR组合模型的丽江国内旅游需求预测  

Forecasting Lijiang Domestic Tourism Demand Based on the GM-ES-GASVR Combination Model

作　　者：党婷[1] 彭乃驰[1] 

机构地区：[1]云南大学旅游文化学院信息科学与技术系,云南丽江674199

出　　处：《数学的实践与认识》2017年第8期279-287,共9页Mathematics in Practice and Theory

基　　金：云南省教育厅科学研究基金(2015Y507)

摘　　要：构建适合于预测丽江国内旅游需求的预测模型,对推动丽江旅游业的发展具有重要意义.研究发现灰色GM(1,1)模型、三次指数平滑模型与GA-SVR模型都适用于预测丽江国内旅游需求,且GA-SVR模型为这三个单项模型中的最优模型.在此基础上,利用变权方法建立GM-ES-GASVR组合预测模型.通过对拟合与测试结果的对比分析,表明GM-ES-GASVR变权组合预测模型比单一模型的拟合与测试效果都有较大改善.Construction of suitable forecasting model for predicting domestic tourism de- mand in Lijiang is of great significance to promote development of Lijiang tourism industry. Research shows that the grey GM（1,1） model, cubic exponential smoothing model and GA- SVR model are suitable for forecasting domestic tourism demand in Lijiang, and the GA-SVR model is the optimal model of the three individual models. The GM-ES-GASVR variable weight combination forecasting model is established on the basis. Through the comparison and analysis of the predicted results, it shows that the forecasting effect of variable weight combination forecasting model is better than the single model.

关 键 词：变权组合预测 灰色GM(1 1) 三次指数平滑 GA-SVR 丽江国内旅游 

分 类 号：F224[经济管理—国民经济] F592.7