基于吉布斯—马尔科夫转换模型的资本资产定价研究

发布时间：2018-03-04 05:20

本文选题：马尔科夫转换模型　切入点：资本资产定价　出处：《宁夏大学》2017年硕士论文　论文类型：学位论文

【摘要】：资本资产定价问题是近年来学术界研究的热点.它不但涉及到传统的投资、理财、保险行业,而且在其他行业也有广泛的应用.传统的研究方法主要是在假定期望收益率服从正态分布的前提下进行的,但在实际操作过程中金融时间序列往往呈现尖峰厚尾或非正态性.本文主要利用基于吉布斯取样的方法构造嵌入马尔科夫转换模型.利用其对动态结构性问题处理的优势,对金融时间序列的时变性进行量化研究,有效地刻画金融时间序列的波动性特征.而后利用吉布斯抽样法构造了求解四区制的马尔科夫转换模型,以上证50股票指数为例进行分析,该模型能较好地刻画金融收益率的波动性特征.实证检验部分利用基于吉布斯—马尔科夫转换的资本资产定价模型对上证300的5只股票进行研究并给出其VaR模型,得出β为系数的驱动因素非一致性和转变时间的不同步性等结论.为投资者进行投资选择和市场监管提供了有效的科学依据.
[Abstract]:The issue of capital asset pricing has been a hot topic in academic circles in recent years. It not only involves the traditional investment, financial management and insurance industries, And it is also widely used in other industries. The traditional research methods are mainly based on the assumption that the expected rate of return is normal distribution. However, the financial time series often show peak and thick tail or non-normality in practice. In this paper, we mainly use Gibbs sampling method to construct embedded Markov transformation model, and make use of its advantage in dealing with dynamic structural problems. In this paper, the time variability of financial time series is quantitatively studied, and the volatility characteristics of financial time series are described effectively. Then, a Markov transformation model for solving four-zone system is constructed by using Gibbs sampling method. Taking the Shanghai Stock Exchange 50 stock index as an example, This model can describe the volatility characteristics of financial yield. The empirical test part uses the capital asset pricing model based on Gibbs-Markov transform to study five stocks of Shanghai Stock Exchange 300 and gives its VaR model. The conclusion that 尾 is the driving factor of the coefficient is inconsistent and the transition time is different, which provides an effective scientific basis for investors to choose their investment and supervise the market.
【学位授予单位】：宁夏大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F224;F832.51

【参考文献】