基于投资者情绪的风险与收益权衡关系实证研究

发布时间：2018-01-22 11:39

本文关键词： 收益风险波动性冲击投资者情绪稳健性　出处：《青岛大学》2013年硕士论文　论文类型：学位论文

【摘要】：股票市场的风险与收益的关系一直是金融理论研究的核心问题,在有效市场假设(EMH)下,传统金融理论的理性资产定价模型(CAPM、APT模型等)认为股票市场的收益是对承担风险的补偿,收益与风险之间应该存在正相关关系,但学者们在过去30多年中对风险与收益权衡关系的实证研究结论并不一致,出现正相关、负相关和关系不显著三种结论。针对这一理论与现实矛盾,从行为金融学理论出发,基于投资者情绪视角对这一问题进行理论和实证研究。首先在阐述风险与收益、波动性冲击关系的理论基础上提出研究假设及推论；然后利用封闭式基金折价、交易量、IPO数目、IPO首日平均收益率、新开户数、好淡指数和消费者信心指数作为投资者情绪代理变量,在剔除宏观经济因素影响后,运用主成分分析方法构建复合投资者情绪指标；进而运用VAR模型和GARCH族模型研究收益与风险、波动性冲击的关系；最后,对比宏观经济变量(利率、GDP增长率和CPI)与复合投资者情绪指标在模型回归结果的正确性来验证复合投资者情绪指标的稳健性,并用复合投资者情绪指标值对收益率进行回归,以进一步证实投资者情绪对收益与风险关系影响的结论。实证分析表明：我国股票市场存在收益与风险显著负相关或正相关不显著现象；投资者情绪高涨不仅会削弱收益与风险的正相关关系,甚至会使其逆转为负相关,但是在投资者情绪低落期,收益与风险是显著正相关的,从而证明以住研究中出现收益与风险负相关或者正相关不显著的原因在于投资者情绪高涨；关于收益与波动性冲击的关系,研究发现,投资者情绪高涨时,收益与波动性冲击正相关,而在投资者情绪低落时,收益与波动性冲击负相关；市场新息和历史信息均对市场波动性产生显著的正向影响,但是在投资者情绪高涨时,新息的正向影响大于投资者情绪低落时期,而历史信息对市场波动性的正向影响则小于投资者情绪低落时期；国内股票市场存在“杠杆效应”——利好消息比利空消息对股票市场波动性冲击效应更大。
[Abstract]:The relationship between risk and return in stock market is always the core problem of financial theory. Under the hypothesis of efficient market, the rational asset pricing model of traditional financial theory is CAPM. The APT model thinks that the return of the stock market is the compensation to bear the risk, and there should be a positive correlation between the return and the risk. However, in the past 30 years, the empirical research conclusions on the trade-off between risk and income are not consistent, there are three conclusions: positive correlation, negative correlation and unsignificant relationship. In view of the contradiction between this theory and reality. Based on the theory of behavioral finance, this paper makes a theoretical and empirical study on this issue from the perspective of investor sentiment. First of all, on the basis of the theory of risk, return, volatility and impact, the research hypothesis and inference are put forward. Then we use closed-end fund discount, the number of IPOs trading volume and IPO average yield, new account number, good light index and consumer confidence index as proxy variables of investor sentiment. After excluding the influence of macroeconomic factors, the main component analysis method is used to construct the composite investor sentiment index. Then we use VAR model and GARCH family model to study the relationship between income, risk and volatility impact. Finally, by comparing macroeconomic variables (interest rate GDP growth rate and CPI) with the correctness of composite investor sentiment indicators in the model regression results, to verify the robustness of composite investor sentiment indicators. In order to confirm the influence of investor sentiment on the relationship between return and risk, the compound investor sentiment index value is used to regression the return rate. The empirical analysis shows that there is a significant negative correlation between return and risk or no significant positive correlation in China's stock market. The upsurge of investor sentiment not only weakens the positive correlation between income and risk, but also reverses the negative correlation. However, in the period of low investor sentiment, income and risk are significantly positively correlated. Therefore, it is proved that the negative correlation between income and risk or the lack of positive correlation is due to the upsurge of investor sentiment. As to the relationship between return and volatility shock, it is found that when investor sentiment is high, return is positively correlated with volatility shock, while when investor is depressed, return is negatively correlated with volatility shock. Both market innovation and historical information have a significant positive impact on market volatility, but when investor sentiment is high, the positive impact of innovation is greater than that of investor depression. The positive impact of historical information on market volatility is smaller than that of investor depression; There is "leverage effect" in domestic stock market.
【学位授予单位】：青岛大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F832.51;F224

【参考文献】