基于搜索大数据的产业需求预测研究

提出引入网络搜索数据指标的动态计量模型研究国际入境旅游需求问题。实证研究数据为面板数据，时间范围为2004年1月至2012年12月，游客来源地为21个中国旅游主要客源国，研究模型为静态及动态GMM模型。研究结果表明，客源地国民生产总值对入境中国旅游起到正相关作用，国民生产越高，入境中国数越多，客源地与人民币汇率与入境数成负相关关系，即汇率越高，入境中国数越少，客源地对中国景区的搜索指数与入境数成正相关关系，搜索指数越高，入境数越高。总之，实证结果表明，客源地GDP、与人民币对入境中国旅游影响显著，与此同时，本入纳入模型的新变量搜索指数对入境中国旅游同样作用显著，因此，在互联网时代，为了提高入境中国旅游人数，除了要关注传播经济指标外，还必须高度关注网络信息建设、信息反馈，因为网络关注度也是一个重要的影响因素。

Research on Industry Demand Forecasting Based on Searching Big Data

This paper presents a dynamic econometrics model to study the demand of international inbound tourism by introducing the index of web search data. The empirical data are panel data. The research models are static and dynamic GMM models. The results show that the gross national product (GDP) of tourists has a positive correlation with the number of tourists to China. The higher the national production is, the more China arrives. The negative correlation between the source nation money and the RMB. The higher the exchange rate , the less inbound China. The search index of tourist attractions in China is related to the number of arrivals. The higher the search index, the higher the number of inbound tourists. In conclusion, the empirical results show that the source GDP and RMB have a significant impact on China''s tourism inbound. In the mean time, the index of new variables incorporated into the model also plays a significant role in China''s inbound tourism. Therefore , In addition to paying attention to the dissemination of economic indicators, China''s tourism population must also pay close attention to the construction of network information and information feedback because the network attention is also an important factor.