投资者情绪对股票市场收益的影响
——基于创业板指数收益率的实证分析
中文摘要
我国始终强调对外开放的大门不会关上，稳定的金融市场是我国产业布局和
经济发展的基础，但通过对比属性相同的创业板市场与纳斯达克市场后发现，我
国创业板指数比美国纳斯达克指数波动幅度更大。基于创业板市场成立时间短、
个人投资者占比大、极易受情绪支配的特点，本文以投资者情绪为视角研究情绪
对创业板指数收益率的影响关系，对保障投资者利益预防证券市场危机具有重要
意义。
国内外学者对投资者情绪研究的关注度较高，在情绪量化过程中多数学者采
用改进量化方法来提高情绪指数精准度，少有学者从指标选取合理性、全面性角
度展开研究。因此，本文在 CICSI情绪指数 6个原始情绪代理指标的基础上，新
增加了投资者信心指数（ICI）、市盈率（PE）、上证指数收益率（SIY）、成
交额（VALUE），共计 10个情绪代理指标，进行两次主成分分析，确定最终投
资者综合情绪指数 ISI2。通过对比 ISI2、CICSI与上证指数、创业板指数的走势
图和相关性，证实了 ISI2具有合理性和更高的准确性。随后本文以创业板指数的
月度收益率作为研究对象，应用 VAR模型、格兰杰因果检验、脉冲响应等方法
分析投资者综合情绪指数 ISI2与创业板指数收益率 R之间的内在关系。
基于以上研究得出结论：（1）投资者情绪与创业板指数收益率互为格兰杰
原因。（2）滞后期投资者情绪对当期投资者情绪的影响表现出持续时间长、衰
退慢的特点。（3）创业板指数收益率对投资者情绪的影响具有滞后性。（4）投
资者情绪是影响创业板指数收益率的主要因素，并且这种影响呈现出先正后负的
现象。为保障投资者利益、稳定创业板市场波动，本文从建立情绪评价体系、加
大风险意识宣传力度、提高监督管理水平、加快注册制推广四个方面提出相应的
对策和建议。
关健词：投资者情绪；创业板指数收益率；主成分分析；VAR模型
I

The Impact of Investor Sentiment on Stock
Market Returns
——An Empirical Analysis Based on the Return
Rate of the Growth Enterprise Market Index
ABSTRACT
Our country has always emphasized that the door of opening to the outside world
will not be closed. A stable financial market is the foundation of my country's
industrial layout and economic development. However, after comparing the Growth
Enterprise Market with the same attributes and the NASDAQ market, we found that
Our country’s Growth Enterprise Market Index is more volatile than the US’s
NASDAQ Composite Index. Based on the short establishment time of the Growth
Enterprise Market, the large proportion of individual investors, and the susceptibility
to emotions, this article uses investor sentiment as a perspective to study the
relationship between emotions and the rate of return of Growth Enterprise Market
Index, which can effectively protect the interests of investors and prevent The stock
market crisis.
Domestic and foreign scholars pay more attention to the research of investor
sentiment. However, in the process of sentiment quantification, most scholars adopt
improved quantification methods to improve the accuracy of sentiment index. Few
scholars conduct research from the perspective of rationality and comprehensiveness
of indicator selection. Therefore, on the basis of the 6 original sentiment proxy
indicators of the CICSI sentiment index, this article adds the investor confidence
index (ICI), the price-earnings ratio (PE), the Shanghai Stock Exchange Index Yield
(SIY), and the turnover (VALUE), a total of 10 The sentiment proxy index is analyzed
by two principal components to determine the final investor's comprehensive
sentiment index ISI2. By comparing ISI2, CICSI, Shanghai Composite Index, Growth
Enterprise Market Index, the trend chart and correlation, it is confirmed that ISI2 is
reasonable and more accurate. Subsequently, this article uses the monthly returns of
II

the Growth Enterprise Market Index as the research object, and analyzes the internal
relationship between the investor's comprehensive sentiment index ISI2 and the
Growth Enterprise Market Index return R through VAR model, Granger causality test,
impulse response and other methods.
Based on the above research, we can draw conclusions: (1) Investor sentiment
and Growth Enterprise Market Index return rate are Granger reasons for each other. (2)
Investor sentiment in the lagging period has a major impact on current sentiment, and
this impact shows the characteristics of long duration and slow recession. (3) The
impact of the Growth Enterprise Market Index yield on investor sentiment is lagging.
(4) Investor sentiment is the main factor that affects the rate of return of the Growth
Enterprise Market Index, and this effect appears to be positive and then negative. In
order to protect the interests of investors and stabilize the volatility of the Growth
Enterprise Market, this article proposes countermeasures and suggestions from four
aspects: establishing an emotional evaluation system, increasing risk awareness
publicity, improving supervision and management, and accelerating the promotion of
the registration system.
Key words: Investor sentiment; Growth Enterprise Market Index Yield;
Principal component analysis; VAR model.
III
。。。以下略