摘要
本文对IC-GARCH模型进行改进,放宽了ICA关于独立成分是IID的假设,考虑独立成分为ARMA模型的平稳过程,提出了基于自相关结构的IC-GARCH估计方法,并给出了估计量的理论性质和计算效率较高的迭代估计算法。最后,对改进方法进行了模拟和实证分析,结果表明本文提出的模型能够使多元GARCH模型应用于高维金融数据,并大大提高了金融资产收益波动率的估计精度。
This paper improves the IC-GARCH model by relaxing the IID assumption of the independent component and considering the independent components as stationary process with ARMA structure. Under a mild assumption, we derive asymptotic properties of the method and establish an efficient iterative algorithm. Finally, we illustrate the proposed procedure can be used in MGARCH model in high-dimensional financial data and demonstrates a very reasonable performance via Monte Carlo simulation and an application to real data.
出处
《数理统计与管理》
CSSCI
北大核心
2017年第1期38-50,共13页
Journal of Applied Statistics and Management
基金
中国人民大学科学研究基金(中央高校基本科研业务费专项资金资助)项目(10XNL007)成果
