Causality results by Granger causality with signal-dependent noise analysis.

The causal influence is from row to column. The causality is given for each direction, and the corresponding -value is presented in brackets. If the uncorrected -value is less than (surviving Bonferroni correction), the causal influence is identified as significant and indicated in bold in the table. (*: The active-set algorithm was used.)

因果影响的方向为行至列。已针对每个方向给出因果关系，对应的p值置于括号内。若未校正的p值小于经邦费罗尼校正（Bonferroni correction）后的显著性阈值，则认定该因果影响具有统计学显著性，并在表格中以加粗字体标注。（*：采用了主动集算法（active-set algorithm）。）