Big data would not lie: prediction of the 2016 Taiwan election via online heterogeneous information

EPJ Data Science

Table 3 Influential significance of events detected for Eric Chu

	Estimate	Std. Error	t value	Pr(>\|t\|)
c	−0.0056	0.0183	−0.307	0.76012
\(p^{llchu}_{k-1}\)	1.0102	0.0481	21.004	<2e–16^∗∗∗
\(\gamma _{9}\)	0.0058	0.0020	2.906	0.0049^∗∗∗
\(\gamma _{10}\)	0.0025	0.0027	0.945	0.3482
\(\gamma _{11}\)	0.0010	0.0024	0.418	0.6776
\(\gamma _{12}\)	0.0023	0.0026	0.871	0.3870
\(\gamma _{13}\)	−0.0002	0.0021	−0.110	0.9130
\(\gamma _{14}\)	0.0049	0.0026	1.842	0.0698^∗
\(\gamma _{15}\)	−0.0262	0.0052	−5.036	3.73e–06^∗∗∗
Residual standard error	0.0050 on 68 degrees of freedom
Multiple R-squared	0.9529
Adjusted R-squared	0.9466
F-statistic	152.7 on 9 and 68 DF	p-value	<2.2e–16

^∗∗∗, ^∗∗, and ^∗ indicate that the coefficients are significantly different from zero at the level of 1%, 5%, and 10%, respectively.