Diagnostics through Residual Plots in Binomial Regression Addressing Chemical Species Data

Table 4

Empirical power of different residuals based on Grubb’s test statistic for outlier’s detection in binomial regression.

		RES	PRES	APRES	CERES (L)	CERES (K)
10	2	0.308	0.610	0.852	0.442	0.650
	10	0.344	0.612	0.872	0.426	0.650
	20	0.314	0.674	0.854	0.428	0.584
	95	0.318	0.650	0.878	0.456	0.612
20	2	0.102	0.742	0.938	0.458	0.866
	10	0.082	0.702	0.926	0.462	0.856
	20	0.124	0.752	0.932	0.422	0.866
	95	0.076	0.734	0.932	0.468	0.870
30	2	0.034	0.762	0.954	0.492	0.944
	10	0.018	0.792	0.936	0.502	0.938
	20	0.026	0.750	0.936	0.522	0.924
	95	0.038	0.776	0.940	0.510	0.936
40	2	0.000	0.786	0.944	0.588	0.996
	10	0.002	0.818	0.948	0.636	0.984
	20	0.002	0.784	0.940	0.586	0.990
	95	0.000	0.796	0.946	0.596	0.990