基于目标设定与自我控制的时间不一致分析框架及其应用

Time inconsistent behavior is common in both real-life and financial applications. As time inconsistency originates in the conflict between long-term interest and short-term interest, the main approach of dealing with time inconsistency in the literature is adopting self-control methods to balance long-term and short-term interests. However, most existing models assume that the objective function of the decision problem is time separable, which limits their applications. Furthermore, the self-control methods in the literature are also limited. This project aims to study the time inconsistency issue by extending existing theoretical models and constructing new research framework. First, we try to extend existing self-control models in order to deal with non-separable decision problems. Second, we try to construct a new self-control model with goal setting, which provides a new angle in dealing with time inconsistency. Third, we will apply both the extended model and the new self-control model to some practical financial decision problems, such as dynamic mean-variance problem, dynamic mean-CVaR problem, and dynamic investment and consumption problem with quasi-hyperbolic discounting time preference. The theoretical results will enrich the study on time inconsistency issue and promote the development of mathematical finance and behavioral finance. The empirical results can help the investors understand and restrict their time inconsistent behaviors in dynamic trading.

时间不一致行为在日常生活和经济金融决策中随处可见。其根源在于决策者长期利益与短期利益之间的矛盾。利用自我控制模型来平衡长短期利益是现有文献处理时间不一致问题的研究主流。然而，现有模型大多假设决策问题的目标函数具有时间可分性，应用范围窄，且自我控制手段有限。本项目拟从扩展现有模型与构建新的研究框架两个方向对时间不一致问题展开系统研究。首先，我们将扩展现有模型，得到适用于一般不可分结构的决策新模型。然后，构建基于目标设定理论的新模型来丰富解决时间不一致问题的手段。最后，应用这两个新框架分析金融实践中常见的时间不一致动态投资组合问题，如动态均值方差问题、动态均值-CVaR问题和具有拟双曲贴现因子的动态投资消费问题等。本项目的理论研究成果能够丰富和发展时间不一致问题的研究内容、理论与方法，推动数理金融和行为金融的发展。本项目的应用研究成果可以帮助投资者了解、约束其在动态投资中的时间不一致行为。

本课题通过借鉴行为科学、管理科学的理论基础，面向经济管理中常见的时间不一致行为，逐渐形成了独具特色的针对时间不一致问题的研究框架和分析体系。主要研究成果及其意义：（1）通过扩展现有的决策者自我序贯博弈模型，构建了适用于一般不可分决策问题的双层博弈模型（Planner-doer model），推进了时间不一致的方法论进展。（2）在构建了双层博弈模型的相关理论之后，将该方法应用于具有拟双曲贴现因子的动态投资消费问题，动态均值方差问题和动态均值-CVaR问题，推进了时间不一致的应用实践。（3）通过构建外部合约的方式约束时间不一致决策者的决策行为，减少时间一致性行为的出现，达到了企业收入和员工福利的双赢。（4）实行合理的自我控制，以及引入适当的竞争均可以较好的处理动态决策问题的时间不一致性，丰富了处理时间不一致的手段。课题负责人共计发表有基金标注的论文8篇，6篇SCI/SSCI检索，1篇EI检索，1篇为受邀综述文章。