投资者情绪与股票收益——来自移动互联网的实证研究
|
摘要
移动互联网的高速发展使得越来越多的投资者通过移动互联网获取信息并做出投资决策。文章利用网络爬虫技术收集来自移动互联网的用户讨论信息,研究来自移动互联网的用户情绪对股票收益的影响,实证结果显示:移动互联网用户情绪存在显著不对称特征,其更倾向于表现积极乐观的情绪,且其正负面情绪差异大于PCs端;同时,移动互联网用户情绪越乐观,下一期股票收益越高。进一步实证结果表明,处于较差信息环境(如散户持股较高,分析师跟踪人数较少)的公司,移动互联网用户情绪对其股票收益的影响更加显著;此外,对于流动性越差的公司,移动互联网用户情绪对其股票收益的影响也越显著。文章研究结论为移动互联网时代的投资者优化投资决策提供了新的视角,也是对行为金融学中传统媒体定价领域的重要补充。
With the rapid development of mobile Internet,more and more investors are obtainimg information and making investment decisions through the mobile Internet. As mobile Internet gaining its popularity in the process of investment,whether and howmobile Internet investor sentiment affects investor behavior as well as stock market performance has become an important unanswered question currently.This paper aims to find out the impact of mobile Internet investor sentiment on stock returns,using the web crawler technology to collect investors' discussion data from mobile Internet. Our dotc comes from Snowball Finance platform( xueqiu. com),one of of the biggest Chinese rnobile Internet stock forums,covering the period from December 1 st,2010 to March 31 st,2015 and April 1 st,2016 to November 31 st,2016. In the first period,all 1100 stocks from Shanghai Stock Exchange are included,while 595 stocks from Shanghai Stock Exchange and Shenzhen Stock Exchange are included in the second period.The empirical results first showthat mobile Internet investors ' sentiment have significant asymmetric characteristics as they are more willing to express positive sentiment. Secondly,comparing to desktop Internet investors,the difference between positive sentiment and negative sentiment is bigger in Internet investors. Thirdly,the more optimistic mobile Internet investors are,the higher next period stock returns. Further empirical results showthat the impact of mobile Internet investor sentiment on stock returns is more pronounced in companies with poor information environments which are characterized with higher retail holdings or fewer analysts. In addition,for companies with less liquidity,the impact of mobile Internet investor's sentiment on stock returns is also more significant.This paper contributes to the literature in several ways. First,by comparing the differences between the discussion information disclosed by mobile and PCs,this paper deepens the understanding of behavior characteristics of mobile Internet investors. Secondly,this paper shows that mobile Internet investors are more willing to express positive emotions,indicating that mobile Internet investors have systematic deviations in stock future earnings expectations,resulting in higher risk premiums for those stocks being paid attention or discussed by mobile Internet investors. Thirdly,this paper explores for the first time the impact of investor sentiment from the mobile Internet on stock returns,enriching the related research and supplementing to the existing media pricing field in behavioral finance.
With the rapid development of mobile Internet,more and more investors are obtainimg information and making investment decisions through the mobile Internet. As mobile Internet gaining its popularity in the process of investment,whether and howmobile Internet investor sentiment affects investor behavior as well as stock market performance has become an important unanswered question currently.This paper aims to find out the impact of mobile Internet investor sentiment on stock returns,using the web crawler technology to collect investors' discussion data from mobile Internet. Our dotc comes from Snowball Finance platform( xueqiu. com),one of of the biggest Chinese rnobile Internet stock forums,covering the period from December 1 st,2010 to March 31 st,2015 and April 1 st,2016 to November 31 st,2016. In the first period,all 1100 stocks from Shanghai Stock Exchange are included,while 595 stocks from Shanghai Stock Exchange and Shenzhen Stock Exchange are included in the second period.The empirical results first showthat mobile Internet investors ' sentiment have significant asymmetric characteristics as they are more willing to express positive sentiment. Secondly,comparing to desktop Internet investors,the difference between positive sentiment and negative sentiment is bigger in Internet investors. Thirdly,the more optimistic mobile Internet investors are,the higher next period stock returns. Further empirical results showthat the impact of mobile Internet investor sentiment on stock returns is more pronounced in companies with poor information environments which are characterized with higher retail holdings or fewer analysts. In addition,for companies with less liquidity,the impact of mobile Internet investor's sentiment on stock returns is also more significant.This paper contributes to the literature in several ways. First,by comparing the differences between the discussion information disclosed by mobile and PCs,this paper deepens the understanding of behavior characteristics of mobile Internet investors. Secondly,this paper shows that mobile Internet investors are more willing to express positive emotions,indicating that mobile Internet investors have systematic deviations in stock future earnings expectations,resulting in higher risk premiums for those stocks being paid attention or discussed by mobile Internet investors. Thirdly,this paper explores for the first time the impact of investor sentiment from the mobile Internet on stock returns,enriching the related research and supplementing to the existing media pricing field in behavioral finance.
引文
郭多祚,徐占东,2002,“中国股票市场Beta和收益关系的实证分析”,《财经问题研究》,第11期,第50-52页。
黄润鹏,左文明,毕凌燕,2015,“基于微博情绪信息的股票市场预测”,《管理工程学报》,第1期,第47-52页。
宋泽芳,李元,2012,“投资者情绪与股票特征关系”,《系统工程理论与实践》,第1期,第27-33页。
Antw eiler W and Frank M Z,2004,“Is all that talk just noise?The information content of internet stock message boards”,The Journal of Finance,59(3):1259-1294.
Amihud Y,2002,“Illiquidity and stock returns:cross-section and time-series effects”,Journal of financial markets,5(1):31-56.
Baker M and Stein J C,2004,“M arket liquidity as a sentiment indicator”,Journal of Financial M arkets,7(3):271-299.
Baker M and Wurgler J,2006,“Investor sentiment and the cross-section of stock returns”,The Journal of Finance,61(4):1645-1680.
B9hmer M,Hecht B and Bauer G,2011,“Falling asleep w ith Angry Birds,Facebook and Kindle:a large scale study on mobile application usage”,Conference on Human-Computer Interaction w ith M obile Devices and Services,M obile Hci 2011,Stockholm,Sw eden,August 30-September.DBLP,47-56.
Bollen J,M ao H and Zeng X,2011,“Tw itter mood predicts the stock market”,Journal of Computational Science,2(1):1-8.
Brow n G W and Cliff M T,2004,“Investor sentiment and the near-term stock market”,Journal of Empirical Finance,11(1):1-27.
Carhart M M,1997,“On persistence in mutual fund performance”,The Journal of finance,52(1):57-82.
Constantiou I D,Lehrer C and Hess T,2014“Changing information retrieval behaviours:an empirical investigation of users’cognitive processes in the choice of location-based services”,European Journal of Information Systems,23(5):513-528.
De Long J B,Shleifer A and Summers L H,et al,1990,“Positive feedback investment strategies and destabilizing rational speculation”,The Journal of Finance,45(2):379-395.
Fama E F and French K R,1993,“Common risk factors in the returns on stocks and bonds”,Journal of financial economics,33(1):3-56.
Fama E F and French K R,2015,“A five-factor asset pricing model”,Journal of Financial Economics,116(1):1-22.
Fang L and Peress J,2009,“M edia coverage and the cross-section of stock returns”,The Journal of Finance,64(5):2023-2052.
Ghose A,Goldfarb A and Han S P,2013,“How Is the M obile Internet Different?Search Costs and Local Activities”,Information Systems Research,24(3):613-631.
Goetzmann W N,Kim D and Kumar A,et al,2015,“Weather-induced mood,institutional investors,and stock returns”,Review of Financial Studies,28(1):73-111.
Hirshleifer D and Shumw ay T,2003,“Good Day Sunshine:Stock Returns and the Weather”,The Journal of Finance,58(3):1009-1032.
Kim,K.Lee,SY.and Kauffman,R.J,2016,“Social sentiment and stock trading via mobile phones”,Proceedings of the 22nd Americas Conference on Information Systems:Surfing the IT Innovation Wave,AM CIS 2016;San Diego,United States,August 11-14.
Koenig-Lew is N,Palmer A and M oll A,2010,“Predicting young consumers'take up of mobile banking services”,International journal of bank marketing,28(5):410-432.
M erton R C,1987,“A simple model of capital market equilibrium w ith incomplete information”,The journal of finance,42(3):483-510.
M ossin J,1966,“Equilibrium in a Capital Asset M arket”,Econometrica,34(4):768-783.
Paredes J C M,M ing C S and M a T,et al,2016,“The influence of smart phone apps on trading behavior”,The Conference on the Theories and Practices of Securities and Financial M arkets.
Sprenger T O,Tumasjan A and Sandner P G,et al,2014,“Tw eets and trades:The information content of stock microblogs”,European Financial M anagement,20(5):926-957.
Solt M E and Statman M,1988,“How useful is the sentiment index?”,Financial Analysts Journal,44(5):45-55.
Yamaguchi T,2006,“Understanding Trading Behavior in 401(K)Plans”,Social Science Electronic Publishing,2006(wp125).
(1)根据2017年第40次中国互联网络信息中心(CNNIC)发布的统计数据,截至2017年6月,我国手机网民规模已达7. 24亿,较2016年底增加2830万人;其中,网民中使用手机上网的人群占比进一步提高,由2016年的95. 1%提高到2017年6月的96. 3%,说明随着移动技术的发展,手机终端的大屏化以及手机应用体验的不断提升,手机作为网民主要上网方式的趋势进一步明显,而通过平板电脑、笔记本电脑或台式电脑接入互联网的比例均有所下降,可见近年来我国移动互联网技术发展快速。
(1)雪球网是一个专业的投资者的社交网络,投资者可以通过雪球网进行跨市场、跨品种的数据查询、新闻订阅和互动交流服务。其中,互动交流是投资者最活跃的版块,截止到2016年10月,雪球用户量已经超过千万,并且雪球用户每天发布30万条UGC内容(User Generated Content,用户原创内容),雪球用户创建的策略组合超过80万个,雪球网用户的发布的数据具有一定的代表性,因此本文选取雪球网作为移动互联网数据的挖掘平台,研究投资者情绪对股票收益的预测作用。
(2)选取周度频率作为本文的研究频率的原因:日度的雪球讨论数据缺失较多,周度频率数据可以使样本区间内的所有样本数据更加统一完整,有助于探索移动互联网信息与股票期望收益的横截面关系;如果采用月度频率,虽然对构建套利策略有利,但是仅仅只有15个月,研究区间较短,所以选取周度频率更加符合本文的研究。
(1)ROST Content Mining(CM)6是武汉大学信息管理学院和计算机学院沈阳教授(现调入清华大学新闻传播学院)团队研发编码的国内以辅助人文社会科学研究的大型免费社会计算平台。该软件可以实现微博分析、聊天分析、全网分析、网站分析、浏览分析、分词、词频统计、英文词频统计、流量分析、聚类分析等一系列文本分析,遍布海内外100多所大学,包括Cambridge University(剑桥大学)、美国Texas A&M University、日本北海道大学、北京大学、清华大学、香港城市大学、澳门大学众多高校,并且已有其他学者近二十篇基于ROST CM的中英文权威或核心级期刊发表。
(2)为了检验ROST软件对帖子情绪测量的有效性,本文随机抽取股票“白云机场(600004)”100条用户帖子,人工判断情绪值,再与ROST软件测量值比较,发现该软件对帖子情绪测量的正确率为63%。
黄润鹏,左文明,毕凌燕,2015,“基于微博情绪信息的股票市场预测”,《管理工程学报》,第1期,第47-52页。
宋泽芳,李元,2012,“投资者情绪与股票特征关系”,《系统工程理论与实践》,第1期,第27-33页。
Antw eiler W and Frank M Z,2004,“Is all that talk just noise?The information content of internet stock message boards”,The Journal of Finance,59(3):1259-1294.
Amihud Y,2002,“Illiquidity and stock returns:cross-section and time-series effects”,Journal of financial markets,5(1):31-56.
Baker M and Stein J C,2004,“M arket liquidity as a sentiment indicator”,Journal of Financial M arkets,7(3):271-299.
Baker M and Wurgler J,2006,“Investor sentiment and the cross-section of stock returns”,The Journal of Finance,61(4):1645-1680.
B9hmer M,Hecht B and Bauer G,2011,“Falling asleep w ith Angry Birds,Facebook and Kindle:a large scale study on mobile application usage”,Conference on Human-Computer Interaction w ith M obile Devices and Services,M obile Hci 2011,Stockholm,Sw eden,August 30-September.DBLP,47-56.
Bollen J,M ao H and Zeng X,2011,“Tw itter mood predicts the stock market”,Journal of Computational Science,2(1):1-8.
Brow n G W and Cliff M T,2004,“Investor sentiment and the near-term stock market”,Journal of Empirical Finance,11(1):1-27.
Carhart M M,1997,“On persistence in mutual fund performance”,The Journal of finance,52(1):57-82.
Constantiou I D,Lehrer C and Hess T,2014“Changing information retrieval behaviours:an empirical investigation of users’cognitive processes in the choice of location-based services”,European Journal of Information Systems,23(5):513-528.
De Long J B,Shleifer A and Summers L H,et al,1990,“Positive feedback investment strategies and destabilizing rational speculation”,The Journal of Finance,45(2):379-395.
Fama E F and French K R,1993,“Common risk factors in the returns on stocks and bonds”,Journal of financial economics,33(1):3-56.
Fama E F and French K R,2015,“A five-factor asset pricing model”,Journal of Financial Economics,116(1):1-22.
Fang L and Peress J,2009,“M edia coverage and the cross-section of stock returns”,The Journal of Finance,64(5):2023-2052.
Ghose A,Goldfarb A and Han S P,2013,“How Is the M obile Internet Different?Search Costs and Local Activities”,Information Systems Research,24(3):613-631.
Goetzmann W N,Kim D and Kumar A,et al,2015,“Weather-induced mood,institutional investors,and stock returns”,Review of Financial Studies,28(1):73-111.
Hirshleifer D and Shumw ay T,2003,“Good Day Sunshine:Stock Returns and the Weather”,The Journal of Finance,58(3):1009-1032.
Kim,K.Lee,SY.and Kauffman,R.J,2016,“Social sentiment and stock trading via mobile phones”,Proceedings of the 22nd Americas Conference on Information Systems:Surfing the IT Innovation Wave,AM CIS 2016;San Diego,United States,August 11-14.
Koenig-Lew is N,Palmer A and M oll A,2010,“Predicting young consumers'take up of mobile banking services”,International journal of bank marketing,28(5):410-432.
M erton R C,1987,“A simple model of capital market equilibrium w ith incomplete information”,The journal of finance,42(3):483-510.
M ossin J,1966,“Equilibrium in a Capital Asset M arket”,Econometrica,34(4):768-783.
Paredes J C M,M ing C S and M a T,et al,2016,“The influence of smart phone apps on trading behavior”,The Conference on the Theories and Practices of Securities and Financial M arkets.
Sprenger T O,Tumasjan A and Sandner P G,et al,2014,“Tw eets and trades:The information content of stock microblogs”,European Financial M anagement,20(5):926-957.
Solt M E and Statman M,1988,“How useful is the sentiment index?”,Financial Analysts Journal,44(5):45-55.
Yamaguchi T,2006,“Understanding Trading Behavior in 401(K)Plans”,Social Science Electronic Publishing,2006(wp125).
(1)根据2017年第40次中国互联网络信息中心(CNNIC)发布的统计数据,截至2017年6月,我国手机网民规模已达7. 24亿,较2016年底增加2830万人;其中,网民中使用手机上网的人群占比进一步提高,由2016年的95. 1%提高到2017年6月的96. 3%,说明随着移动技术的发展,手机终端的大屏化以及手机应用体验的不断提升,手机作为网民主要上网方式的趋势进一步明显,而通过平板电脑、笔记本电脑或台式电脑接入互联网的比例均有所下降,可见近年来我国移动互联网技术发展快速。
(1)雪球网是一个专业的投资者的社交网络,投资者可以通过雪球网进行跨市场、跨品种的数据查询、新闻订阅和互动交流服务。其中,互动交流是投资者最活跃的版块,截止到2016年10月,雪球用户量已经超过千万,并且雪球用户每天发布30万条UGC内容(User Generated Content,用户原创内容),雪球用户创建的策略组合超过80万个,雪球网用户的发布的数据具有一定的代表性,因此本文选取雪球网作为移动互联网数据的挖掘平台,研究投资者情绪对股票收益的预测作用。
(2)选取周度频率作为本文的研究频率的原因:日度的雪球讨论数据缺失较多,周度频率数据可以使样本区间内的所有样本数据更加统一完整,有助于探索移动互联网信息与股票期望收益的横截面关系;如果采用月度频率,虽然对构建套利策略有利,但是仅仅只有15个月,研究区间较短,所以选取周度频率更加符合本文的研究。
(1)ROST Content Mining(CM)6是武汉大学信息管理学院和计算机学院沈阳教授(现调入清华大学新闻传播学院)团队研发编码的国内以辅助人文社会科学研究的大型免费社会计算平台。该软件可以实现微博分析、聊天分析、全网分析、网站分析、浏览分析、分词、词频统计、英文词频统计、流量分析、聚类分析等一系列文本分析,遍布海内外100多所大学,包括Cambridge University(剑桥大学)、美国Texas A&M University、日本北海道大学、北京大学、清华大学、香港城市大学、澳门大学众多高校,并且已有其他学者近二十篇基于ROST CM的中英文权威或核心级期刊发表。
(2)为了检验ROST软件对帖子情绪测量的有效性,本文随机抽取股票“白云机场(600004)”100条用户帖子,人工判断情绪值,再与ROST软件测量值比较,发现该软件对帖子情绪测量的正确率为63%。