Financial Innovation

Table 2 The results of the power-law distribution test under different manufactured thresholds, Δ

From: Detecting the lead–lag effect in stock markets: definition, patterns, and investment strategies

Δ	CSI 300						S&P 500
Δ	\(\hat{\lambda }\) of K–S	p value	\(\hat{\lambda }\) of Kuiper	p value	\(\hat{\lambda }\) of A–D	p value	\(\hat{\lambda }\) of K–S	p value	\(\hat{\lambda }\) of Kuiper	p value	\(\hat{\lambda }\) of A–D	p value
0.10	− 10.946	0.088	− 13.524	0.999	− 12.684	0.097	− 19.124	0.466	− 11.476	0.999	− 11.476	0.127
0.15	− 15.844	0.392	− 16.582	0.999	− 18.298	0.427	− 29.637	0.998	− 20.201	0.999	− 25.722	0.364
0.20	− 20.241	0.460	− 16.577	0.999	− 20.241	0.458	− 28.584	0.993	− 23.252	0.999	− 27.778	0.752
0.25	− 24.245	0.999	− 16.510	0.999	− 20.301	0.686	− 30.000	0.994	− 26.959	0.999	− 30.000	0.719
0.30	− 27.579	0.927	− 18.807	0.999	− 24.206	0.506	− 21.927	0.962	− 29.009	0.999	− 23.228	0.654

The statistical significance level is set to 0.05. Here, \(\hat{\lambda }\) is the estimated value of parameter λ from the probability function of the power-law distribution (i.e., \(p(x = k) \propto k^{ - \lambda }\))

Back to article page