基于分位数因子模型的高维时间序列因果关系分析
|
|
梁慧玲, 刘慧, 刘力维, 赵佳, 阮怀军
|
Causal relationship analysis of high⁃dimensional time series based on quantile factor model
|
|
Huiling Liang, Hui Liu, Liwei Liu, Jia Zhao, Huaijun Ruan
|
|
表2 VAR10(5)的100次蒙特卡洛实验中因果关系结果的频率
|
Table 2 The frequency of causality results in 100 Monte Carlo implementations of VAR10(5)
|
|
| 方法 | CGC | PMIME | PCA⁃CGC | mBTS⁃CGC | QFM⁃CGC |
|---|
| X1→X2 | 99% | 99% | 100% | 99% | 100% | | X1→X6 | 32% | 95% | 75% | 57% | 91% | | X1→X9 | 35% | 98% | 40% | 55% | 100% | | X2→X4 | 0 | 100% | 85% | 99% | 100% | | X2→X5 | 0 | 1% | 0 | 0 | 0 | | X4→X5 | 88% | 99% | 0 | 99% | 100% | | X5→X1 | 51% | 85% | 100% | 57% | 100% | | X5→X3 | 36% | 100% | 100% | 99% | 100% | | X5→X4 | 95% | 69% | 100% | 97% | 100% | | X5→X8 | 23% | 99% | 100% | 80% | 94% | | X6→X7 | 0 | 100% | 99% | 89% | 100% | | X6→X8 | 0 | 0 | 0 | 0 | 8% | | X7→X10 | 100% | 100% | 100% | 99% | 100% | | X8→X6 | 100% | 96% | 100% | 99% | 100% | | X9→X8 | 0 | 1% | 0 | 0 | 0 | | X10→X8 | 0 | 1% | 0 | 0 | 0 | | X10→X9 | 100% | 100% | 100% | 99% | 100% |
|
|
|