基于特质风险的股票市场投资策略研究

发布时间：2018-09-11 16:45

【摘要】：传统的投资组合理论和CAPM模型认为，特质风险可以通过分散化投资而规避掉，然而，现实中由于交易成本、风险偏好等因素的影响，投资者并无法进行很好的分散化投资，从而不得不承担特质风险。本文正是在这样的背景下，从特质风险问题出发，在系统归纳和总结关于特质风险的理论和实证研究的基础上，通过研究特质风险与预期收益、投资者行为的关系，结合行为金融学的相关结论，构建了基于特质风险的股票市场投资策略，即低特质风险的股票采取动量策略，高特质风险的股票采取反转策略。本文得到主要研究结论如下：首先，本文研究了特质风险与预期收益的关系。通过公式和模型推导得出基于特质风险的定价模型，模型表明股票的预期收益不仅受系统性风险的影响，还受特质风险的影响，从理论上证明了特质风险可以得到定价。然后，本文采用能够刻画波动率时序特征的EGARCH模型来估计特质风险，以横截面回归分析和投资组合分析为实证研究方法，研究得出特质风险与预期收益之间存在显著的正相关关系，且在不同的时间区间都稳定的存在，研究表明我国股市并不存在所谓的“特质波动率之谜”，造成这种异象的根本原因在于忽视了特质波动率的时间序列属性。其次，本文研究了机构投资者行为对于特质风险及其风险溢酬的影响。本文通过实证研究发现，机构的大量持股有助于降低股票的特质风险，机构持股比例增加，股票的特质波动率减少，，机构投资者持股比例越低的股票，特质风险越大，股票预期收益越多。文章从机构持股的角度分析和探讨特质风险与预期收益的关系，进一步指出投资分散化的重要性，并认为投资者无法充分分散化投资是导致特质风险正向定价的原因。最后，本文结合行为金融学的相关理论和特质风险的相关结论，构建了基于特质风险的股票市场投资策略。通过基于特质风险的动量效应和反转效应分析，发现特质风险越高的股票，其反转效应越明显，因此得出这样一个投资策略，即对低特质风险的股票采取动量策略，对高特质风险的股票采取反转策略，经过历史数据测试后，能够明显的跑赢基准指数。
[Abstract]:According to traditional portfolio theory and CAPM model, trait risk can be avoided by diversification. However, due to the influence of transaction cost, risk preference and other factors, investors can not make a good diversification investment. So they have to take the risk of idiosyncrasies. It is under this background that this paper starts from the problem of trait risk, on the basis of systematically summarizing and summarizing the theoretical and empirical research on trait risk, through the study of the relationship between trait risk and expected income, investor behavior, and so on. Based on the relevant conclusions of behavioral finance, this paper constructs a stock market investment strategy based on trait risk, that is, stocks with low trait risk adopt momentum strategy and stocks with high trait risk adopt reverse strategy. The main conclusions are as follows: firstly, this paper studies the relationship between trait risk and expected return. The pricing model based on idiosyncratic risk is derived by formula and model. The model shows that the expected return of stock is affected not only by systematic risk, but also by idiosyncratic risk, which proves theoretically that idiosyncratic risk can be priced. Then, we use EGARCH model, which can describe volatility time series, to estimate trait risk. Cross section regression analysis and portfolio analysis are used as empirical research methods. The study shows that there is a significant positive correlation between trait risk and expected return, and it is stable in different time intervals. The study shows that there is no so-called "trait volatility puzzle" in China's stock market. The fundamental reason for this anomaly lies in the neglect of the time series attribute of the idiosyncratic volatility. Secondly, this paper studies the influence of institutional investor behavior on trait risk and risk overpayment. Through empirical research, this paper finds that a large number of institutional holdings can help to reduce the specific risk of the stock, the increase of institutional ownership ratio, the decrease of the volatility of the stock, the lower the proportion of institutional investors, the greater the specific risk. The more stocks are expected to return. This paper analyzes and discusses the relationship between trait risk and expected income from the perspective of institutional shareholding, further points out the importance of diversification, and points out that investors' inability to fully diversify their investment is the reason leading to positive pricing of idiosyncratic risk. Finally, based on the theory of behavioral finance and the conclusion of trait risk, this paper constructs a stock market investment strategy based on idiosyncratic risk. Through the analysis of momentum effect and reversal effect based on trait risk, it is found that the higher the trait risk is, the more obvious the reverse effect is. Adopting reverse strategy for high trait risk stocks can outperform benchmark index obviously after historical data test.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F832.51;F224

【参考文献】