Granger因果分析模型研究进展

Review on the Progress for Modeling Granger Causality Analysis

由于时间序列的非平稳性,不具有有限方差,所以高斯—马尔科夫定理不再成立,用普通最小二乘法得到的参数估计不再是一致的,出现伪回归的现象,从而导致错误的因果关系。同时由于时间序列的非线性,常规的线性向量自回归模型难以正确描述经济变量之间的因果关系。从G ranger因果分析模型研究进展情况,可见目前所面临的问题和未来可能的发展方向。

Because the non-stationary time sequence don＇t have limited variance and it cannot accord with Gauss- Markov Theorem, the Ordinary Least-Squares Estimators are inconsistent and then the spurious regression occurs, thus the incorrect causality can be drawn. At the same time, due to the non-linear condition of time sequence, conventional linear Vector Autoregressive model can hardly characterize the causality among economic variables correctly. Review on the progress for modeling Granger causality analysis can indicate existing problems and possible direction of future development.