网络搜索对股票市场的预测能力:理论分析与实证检验  

Predictive Power of Internet Search Data for Stock Market:A Theoretical Analysis and Empirical Test

作　　者：刘颖[1] 吕本富[1] 彭赓[1] 

机构地区：[1]中国科学院研究生院管理学院,北京100190

出　　处：《经济管理》2011年第1期172-180,共9页Business and Management Journal ( BMJ )

基　　金：国家自然科学基金"信息不对称程度的量化研究--基于电子商务市场交易数据的实证分析"(70972104);国家自然科学基金"电子商务交易动力机制研究"(70772103);北京市自然科学基金"北京市电子政务信息资源共建共享投资回报研究"(9073020);阿里巴巴青年学者支持计划"网络购物用户行为和购买力提升研究"(Ali-2010-A-5)

摘　　要：网络搜索数据记录了数以亿计的搜索关注与需求,为研究市场交易行为提供了必要数据基础。本文以股票市场为例,首先从微观的投资者行为视角建立一个理论框架,揭示了网络搜索与股票市场之间存在一定的先行——滞后关系。然后,在时差相关分析的基础上,根据经济含义将搜索数据合成为三类搜索指数:股民行动指数、市场行情指数、宏观形势指数。实证检验得出,搜索指数与上证指数年收益率正相关且存在协整关系。在长期趋势中,三类搜索指数分别每增加1个百分点,年收益率将增加0.22、0.56、0.83个百分点。进一步的Granger因果关系检验表明,搜索指数对上证指数年收益率具有显著的预测能力。The Internet search data, recording hundreds of millions searchers＇ concerns and intends, can provide necessary data basis for the study of market transaction behaviors. In this paper, taking the stock market as a research object, we firstly establish a theoretical framework to reveal a lead-lag relationship between search data and stock market based on micro-perspective of investors＇ behaviors. Then according the meanings of search key- words and cross-correlation analysis, we develop three types of composite search index ： investor action index, mar- ket condition index, and macroeconomic index. The empirical test indicates cointegration relationship between search index and the annualized yield of Shanghai composite index. In the long-term trend, each additional one percentage point increase in the three types search indices separately, the annualized yield will increase 0. 22, 0.56,0.83 percentage points in the next month. Furthermore, Granger causality test shows that the search indices have significant predictive power for the annualized yield of Shanghai composite index.

关 键 词：网络搜索数据 股票市场 时差相关 协整分析 GRANGER因果关系 

分 类 号：F830.91[经济管理—金融学]