Skip to main content

Table 6 Causal Links: IRA. We show the value of the maximal causal effect, \(I_{i \to j}^{{\mathrm{CE}, \max}} = \max_{0 < \tau \le \tau _{\max}} \vert {I_{i \to j}^{{\mathrm{CE}}}(\tau )} \vert \) between each pair \((i, j)\) of activity time series, where \(\tau _{\max}= 18 \times 15\) min = 4.5 h is the maximal time lag considered, with standard errors. The arrows indicate the direction of the causal effect. For each activity time series, we indicate in bold the most important drivers of activity (excluding themselves). In blue, we highlight the auto-correlation of each node

From: Suspended accounts align with the Internet Research Agency misinformation campaign to influence the 2016 US election