基于改进一般失望模型的投资组合与资产定价研究

发布时间：2018-01-29 23:19

本文关键词： 一般失望模型投资组合资产定价风险偏好定价因子　出处：《哈尔滨工业大学》2013年博士论文　论文类型：学位论文

【摘要】：经典投资组合和资产定价理论采用了源于新古典主义经济学的同质经济概念，假设投资者的经济决策是完全理性的。但理性并非人类行为的唯一驱动力，还会更多地受主观偏好等多重因素的影响。对资产的收益和风险采用不同的度量方法，会得到不同的投资组合方案，进一步也会得出不同的资产定价模型。投资者所持有的资产具有不同的类型和不同程度的风险，这些风险会给投资者带来相应的损失，因此投资者会对相应的资产进行必要的与风险相关的分析和评估，，从而采取符合其投资目的和风险偏好的投资策略，减少其可能遭受的损失。而资产定价模型反应的则是整个市场处于均衡状态时资产的风险与其均衡价格之间的关系。因此深入研究并理解已有先进投资组合与资产定价理论的相关研究成果，在此基础上根据行为主体的真实心理感受及市场的实际运行状况对相关理论进行发展与完善，具有重要的理论与实际意义。本文首先对已有的风险偏好和风险度量以及投资组合优化和资产定价理论进行了总结与评价，指出已有理论存在的缺点与不足。在此基础上对已有理论进行改进与完善，提出了改进的一般失望模型，在这一改进的模型中，原有的行为主体目标收益参考点由固定的期望收益变为更为灵活的一般性收益参考点，同时采用适用性更强的线性分段效用函数替代了原有的效用函数，使得新的改进后的一般失望模型能够反映行为主体的多层次风险感受。其次，本文在改进的一般失望模型基础上构建了两类新的投资组合优化模型，其中一类以改进的一般失望模型作为目标函数，这类投资组合优化模型能够更为全面的反映行为主体或投资者的多重风险偏好，并为其提供相应的更为符合要求的投资组合方案，另一类投资组合优化模型以改进的一般失望模型对投资组合收益的偏度进行控制。与已有的偏度控制方法相比，以改进的一般失望模型进行偏度控制可以在满足相同投资目标的情形下为投资者提供更多的获取超额收益的机会。再次，本文在基于改进的一般失望模型的投资组合优化模型的基础上，推导出了单参考点情形下的资产定价模型，进而又给出了多参考点情形下的资产定价模型，并对这一模型进行了相关的实证研究，同时对模型中不同风险偏好对应的不同贝塔系数进行时变特征的研究，以系统的分析不同风险偏好对应的不同类型资产定价因子的时变特征。最后，本文对多参考点情形的基于改进一般失望模型的资产定价模型进行了进一步扩展，使其包括更多类型的资产定价风险因子，并以这个模型为基础，将流动性风险因子引入到资产定价模型中，同时本文对给出的包含流动性风险定价因子的定价模型进行相关实证研究，并进一步分析流动性风险贝塔的各种时变性特征，研究结果表明，本文对资产定价模型进行的扩展性研究是合理可行的。
[Abstract]:The classic portfolio and the theory of asset pricing adopt the homogeneous economic concept derived from the neo - classical economics . It is assumed that the investor ' s economic decision is totally rational . However , it is not the only driving force for the human behavior , but it can be more influenced by multiple factors , such as subjective preference . This paper summarizes and evaluates existing risk appetite and risk measurement and portfolio optimization and asset pricing theory , and points out the shortcomings and shortcomings of the existing theory . In this improved model , the original target revenue reference point is changed into more flexible general revenue reference point by fixed expected return , and the original utility function is replaced by the linear segmentation utility function with stronger applicability , so that the new improved general disappointment model can reflect the multi - level risk feeling of the actor . Secondly , based on the improved general disappointment model , two kinds of new portfolio optimization models are constructed . One kind of investment portfolio optimization model can reflect the multiple risk preference of the actor or investor more comprehensively , and provide the corresponding more satisfactory investment combination scheme . The other kind of portfolio optimization model can provide the corresponding more satisfactory investment combination scheme , and another kind of portfolio optimization model can provide the corresponding more satisfactory investment combination scheme , and the other investment portfolio optimization model can provide more opportunities for investors to obtain excess earnings in the case of meeting the same investment objectives . Thirdly , based on the investment portfolio optimization model based on the improved general disappointment model , the asset pricing model in the case of single reference point is derived , then the asset pricing model in the case of multiple reference points is presented , and the relevant empirical research is carried out on the model , meanwhile , the time - varying characteristics of different Beta coefficients corresponding to different risk preferences in the model are studied , and the time - varying characteristics of different types of asset pricing factors corresponding to different risk preferences are analyzed in the system . Finally , this paper further expands the asset pricing model based on the improved general disappointment model of the multi - reference point situation , which includes more types of asset pricing risk factors and introduces the liquidity risk factors into the asset pricing model on the basis of this model .

【学位授予单位】：哈尔滨工业大学
【学位级别】：博士
【学位授予年份】：2013
【分类号】：F830.59

【参考文献】