|
|||||
|
|
| 基于“谷歌趋势”数据的入境外国游客量预测 | |
| 引用本文： | 沈苏彦,赵锦,徐坚.基于“谷歌趋势”数据的入境外国游客量预测[J].资源科学,2015,37(11):2111-2119. |
| 作者姓名： | 沈苏彦 赵锦 徐坚 |
| 作者单位： | 1. 南京林业大学旅游管理系,南京 210037 2. 南京信息职业技术学院信息服务学院,南京 210023 |
| 基金项目： | “基于恢复力理论的旅游历史街区演化过程及机制研究”（41301150） |
| 摘 要： | 入境游客量的预测是制定旅游发展规划和相关政策法规的重要依据。基于“谷歌趋势”提供的涉及旅游活动食、住、行、游、购、娱等环节的相关关键词搜索数据,通过计算相关系数,找出与国家旅游局公布的2004年1月至2015年3月中国入境外国游客量统计数据密切相关的搜索关键词。同时,利用2004年1月至2012年12月的入境外国游客量数据构建一般季节性乘积ARIMA模型,以及带搜索关键词作为自变量的季节性乘积ARIMA模型,分别对2013年1月至2015年3月入境外国游客量进行模拟预测,比较两模型的拟合程度和预测能力。研究发现：加入谷歌关键词作为自变量的季节性乘积ARIMA模型比一般季节性乘积ARIMA模型拟合效果和预测精度高,而中国签证政策与航班信息均对入境外国游客量有显著的影响。 |
| 关 键 词： | 谷歌趋势 入境外国游客 游客量 ARIMA模型 |
| 收稿时间： | 2015-07-10 |
| 修稿时间： | 2015-10-15 |
Forecasting China inbound tourism with google trend data |
|
| SHEN Suyan,ZHAO Jin,XU Jian.Forecasting China inbound tourism with google trend data[J].Resources Science,2015,37(11):2111-2119. | |
| Authors: | SHEN Suyan ZHAO Jin XU Jian |
| Institution: | 1. Tourism Department,Nanjing Forestry University,Nanjing 210037,China 2. Information Service Department,Nanjing College of Information Technology,Nanjing 210023,China |
| Abstract: | Forecasting inbound tourism is very important for destination nations’ tourism development planning and related development policies. Search engine query data related to tourism activities in China （e.g. fights to China,hotels in China,restaurants in China,shopping in China,and destinations in China） were collected from Google Trend. We examined the relationship between search engine query data and inbound foreign visitor volumes to China using Google Trend data and real visitor data from January 2004 to March 2015 provided by the China National Tourism Administration. According to Pearson correlation coefficients,inbound foreign visitor volumes to China have a relatively strong positive relationship with three search engine queries：restaurants in China,flights to China and visas for China. A seasonal ARIMA model without search engine data and a seasonal ARIMA model with search engine data were constructed based on data from January 2004 to December 2012. Two models were compared via model fit statistics and forecast capability based on statistical data from January 2013 to March 2015. We found that the seasonal ARIMA model with search engine data was better than the model without search engine data regarding model fit statistics and prediction accuracy. This study demonstrates that Google Trend can be used for forecasting inbound tourism demand and that flights to China and visas to China are two significant factors influencing inbound tourism volumes. |
| Keywords: | Google trend data China inbound visitor visitor volumes ARIMA model |
| 点击此处可从《资源科学》浏览原始摘要信息 | |
| 点击此处可从《资源科学》下载免费的PDF全文 | |