金融市场的风险控制指标和模型研究

发布时间：2018-01-12 10:14

本文关键词：金融市场的风险控制指标和模型研究　出处：《广西大学》2014年硕士论文　论文类型：学位论文

【摘要】：从Markowitz第一次使用方差和协方差测度风险以来,学者们对金融市场的风险控制指标进行了大量的研究.半个多世纪以来,研究者们先后提出了各种各样的风险控制指标,比如半方差,绝对离差,半绝对离差,在险价值,条件在险价值等.并构建含有这些风险控制指标的投资组合模型,同时设计更快捷有效的计算方法.但是,无论使用哪种风险控制指标,都存在着这样或那样的不足.为了弥补这些风险控制指标的不足,理论界展开了广泛的研究和探讨.论文就是对风险控制指标和模型做进一步的讨论,主要内容和成果如下：为了更好的控制风险,文章在负绝对离差风险控制指标中,考虑偏离程度的波动性,构建了含有波动性的负绝对离差风险控制指标.并将此波动性引入正绝对离差风险控制指标.然后通过调整因子将含有波动性的正绝对离差指标和含有波动性的负绝对离差指标相结合,构建新的风险控制指标.在此指标下建立投资组合模型,并运用混合遗传算法来求解模型并给出具体算例. 给出一致性风险控制指标CVaR的定义,分析其数量经济学意义.对含有此风险控制指标的模型分别进行线性化和离散化处理,建立单损失的MCVaR模型.进一步考虑市场的摩擦因子,在单损失模型中添加改进的典型交易成本函数.然后在相关性粒子群算法中引入动态下降的惯性权重,用此改进后的算法来求解模型并给出具体算例.
[Abstract]:Since Markowitz first used variance and covariance to measure risk, scholars have done a lot of research on the risk control index of financial market for more than half a century. Researchers have proposed a variety of risk control indicators, such as semi-variance, absolute deviation, semi-absolute deviation, at risk value. A portfolio model with these risk control indicators is constructed, and a faster and more effective calculation method is designed. However, no matter which risk control index is used. In order to make up for the shortcomings of these risk control indicators, the theoretical circle has carried out a wide range of research and discussion. This paper is to further discuss the risk control indicators and models. The main elements and outcomes are as follows: In order to control the risk better, the volatility of deviation degree is considered in the negative absolute deviation risk control index. The negative absolute deviation risk control index with volatility is constructed, and the volatility is introduced into the positive absolute deviation risk control index. Then, the positive absolute deviation index with volatility and the positive absolute deviation index with volatility are adjusted by adjusting factors. The negative absolute deviation index is combined. A new risk control index is constructed, under which a portfolio model is established, and a hybrid genetic algorithm is used to solve the model and an example is given. The definition of consistent risk control index (CVaR) is given, and its quantitative economic significance is analyzed. The model with this risk control index is linearized and discretized, respectively. The MCVaR model of single loss is established and the friction factor of the market is further considered. The improved transaction cost function is added to the single loss model, and then the dynamic decreasing inertia weight is introduced into the correlation particle swarm optimization algorithm. The improved algorithm is used to solve the model and an example is given.
【学位授予单位】：广西大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F224;F832.5

【参考文献】