论文部分内容阅读
在无风险资产和有风险证券的离散时间资产定价问题中,常用包含相关的随机成分和非随机成分的增量过程模型来表示.受此启发,文章提出了一类融合了非随机和随机成分的半参数回归模型.与经典的回归模型不同,在此模型中均值回归函数包含了方差部分,并且模型变量与某个状态变量有关联,因此模型更具有特定的经济意义.文中的一个例子解释了GARCH-M模型与现有的广义漂移模型不能包含本文中所提出的模型.文章还表明,虽然增量过程只是两个部分的加权和,但模型的统计推断不能够简单地通过两个独立系统来完成.文章研究了估计量的渐近理论性质,并通过蒙特卡洛模拟考察了估计量的小样本性质.最后利用中国金融年鉴2004-2005的数据分析了中国金融市场的财富增量过程.
In the discrete-time asset pricing of risk-free and risk-based securities, an incremental process model with related stochastic components and non-stochastic components is often used to represent it.On this basis, the paper proposes a class of non-stochastic and stochastic components The semi-parametric regression model is different from the classical regression model, in which the mean regression function contains the variance part, and the model variable is associated with a state variable, so the model has more specific economic significance. The GARCH-M model and the existing generalized drift model can not contain the model proposed in this paper.The article also shows that although the incremental process is only a weighted sum of the two parts, the statistical inference of the model can not be easily achieved by two independent System.This paper studies the asymptotic properties of estimators and investigates the small sample size of the estimators by Monte Carlo simulation.Finally, we use the data from China Financial Yearbook 2004-2005 to analyze the wealth increment process in China’s financial markets .