【作者】 张敏华；

【导师】 张小涛；

【作者基本信息】 天津大学 ， 金融硕士（专业学位）， 2021， 硕士

【摘要】 羊群效应对股票市场的稳定性具有重要影响,它不仅是金融异常现象之一,在行为金融研究领域占据重要地位。随着现代信息技术的飞速发展,互联网信息平台已经成为投资者获取股票信息的主要渠道之一。东方财富证券吧是互联网信息平台的代表。所以,利用东方财富网吧的发帖数据,研究互联网信息交换平台对股市行业羊群效应的影响,具有一定的学术意义和实践意义。本文主要研究了行业发帖数量对行业羊群效应的影响,文中选取2008-2019年的东方财富股吧发帖数据以及国泰安数据库中的羊群效应的横截面绝对偏差CSAD数据,研究行业发帖数量对行业羊群效应的影响,分析行业发帖数量对行业羊群效应的调节作用。本文的结论主要有以下三条:第一,行业股吧发帖数量对行业羊群效应具有显著影响。羊群效应与行业发帖量正相关,即行业发帖量越多,羊群效应越显著,因为人们对消息趋之若鹜。第二,行业股吧发帖交流中的情绪因素对行业羊群效应的影响具有不对称性。对股吧发帖内容根据情感倾向分组后进行分组回归,发现帖子的情感倾向不同时,行业羊群效应的程度也不相同,具体表现为发帖情感正向行业羊群效应<发帖情感中性行业羊群效应<发帖情感负向行业羊群效应。在调节效应方面,带有中性情感或者带有负面情感的行业发帖对羊群效应的影响大于带有正面情感的行业发帖对羊群效应的影响。第三,市场上存在行业羊群效应,同时,在各个行业内部的行业羊群效应大小存在差异,其中住宿和餐饮行业(即行业H)的羊群效应最大;农林牧副渔业、交通运输仓储和邮政业、文化体育和娱乐业次之;其他行业的羊群效应均在2.5左右。更多还原

【Abstract】 Herding has a considerable influence on the stability of the stock market.It plays an important role in the research field of behavioral finance.With the rapid development of modern information technology,Internet information platforms have become one of the main channels for investors to obtain stock information.Oriental Fortune Securities Bar is a representative of the Internet information platform.Therefore,using the posting data of CSMAR database to study the impact of Internet information exchange platform on herding behavior in securities market is worthwhile.This article mainly studies the influence of the number of industry posts on the industry herding effect.The paper selects the 2008-2019 Eastern Fortune Stock Bar’s posting data and the cross-sectional absolute deviation CSAD data of the herding effect in the CSMAR database to study the impact of the number of industry posts on the industry.In terms of the impact of herding,this paper analyzes the regulatory effect of the number of industry posts on industry herding.The conclusions of this article are mainly as follows:First,the number of posts posted by industry stock bars has a significant impact on the industry herd effect.The herd effect is positively correlated with the number of posts posted by the industry,that is,the more posts posted by the industry,the more significant the herd effect.Second,the influence of emotional factors in the post exchanges of industry stock bars on the industry herd effect is asymmetry.The content of the posts on the stock bar is grouped according to the sentiment tendency and then the group regression is carried out.It is found that the degree of the industry herd effect is also different when the sentiment tendency of the post is different.Regarding the moderating effect,industry posts with neutral or negative emotions have a greater impact on the herd effect than industry posts with positive emotions have on the herd effect.Third,there is industrial herding in the market.In addition,there are differences in the size of industrial herding in each industry.更多还原