基于投资者关注的股价走势预测与交易策略设计——股票间交叉模式视角

股票搜索量作为投资者关注的重要代理变量,其变动情况与该股票价格走势的关系得到了学界与业界的广泛关注.另一方面,股票间的价格同步性一直是股票市场中人们重点关注的话题.基于此,本文在个股层面探究了相关股票搜索量变动与目标股票价格走势的关系模式,利用延迟时态关联分析提出了发现股价预测规则的方法,并进一步设计了个股层面的交易策略.同时,本文以我国A股市场股票为研究样本,针对特定场景检验了个股层面交易策略的效果.实验结果表明:利用相关股票搜索量与目标股票价格的关系,可以获得显著的超额收益.

Stock price prediction and trading strategy design based on investors' attention: Cross patterns perspective between stocks

With the rise of search engines, stock search volume has been well recognized as an important indicator of investor attention. In addition, stock price synchronicity also attracts many researchers' attention. To jointly consider these two phenomenons, this paper analyses the relationship between search volume of relevant stock and stock price. Unlike previous research that discusses the relationship at market level, this paper analyses the relationship at individual stock level. An after temporal associative analysis is applied to discover patterns between relevant stock volume variations and subsequent stock price changes, and trading strategy is further designed. To verify the proposed trading strategy, experiments are conducted on China A-share stocks under specific experimental setups. Results show that trading strategies which consider the relationship between search volumes and prices perform better than strategies without considering the relationship. Further, stock level strategy performs better than market level strategy.