Development of New Robust Optimal Score Function for the Weibull Distributed Error Term in Multilevel Models

Table 13

Bias and MSE of using Weibull, Wilcoxon, and REML methods when  = 3..

		REML		Weibull		Wilcoxon		Precision
N2	N1	Bias	MSE	Bias	MSE	Bias	MSE	Weib/REML	Weib/Wil	Wil/REML
5	5	2.085	2.459	0.197	0.521	2.115	2.692	1.057	0.965	1.095
	10	2.066	3.441	0.23	0.754	2.066	3.603	0.998	0.953	1.047
	20	2.099	5.771	0.237	1.257	2.106	5.955	0.99	0.96	1.032
	40	2.102	10.379	0.208	1.9	2.115	10.554	0.967	0.951	1.017
10	5	2.205	1.927	0.119	0.344	2.203	1.984	1.013	0.984	1.03
	10	2.208	2.569	0.103	0.381	2.201	2.612	0.995	0.979	1.017
	20	2.22	3.842	0.089	0.507	2.215	3.874	0.987	0.979	1.008
	40	2.216	6.396	0.1	0.895	2.209	6.429	0.985	0.98	1.005
20	5	2.275	1.701	0.029	0.21	2.265	1.707	0.994	0.991	1.004
	10	2.276	2.007	0.04	0.265	2.266	2.004	0.989	0.99	0.999
	20	2.259	2.634	0.038	0.294	2.248	2.612	0.986	0.994	0.992
	40	2.268	3.99	0.039	0.506	2.256	3.954	0.983	0.992	0.991
40	5	2.279	1.509	0.037	0.205	2.256	1.502	0.993	0.997	0.995
	10	2.284	1.682	0.014	0.201	2.266	1.666	0.989	0.998	0.99
	20	2.307	2.054	0.004	0.248	2.293	2.024	0.984	0.999	0.985
	40	2.291	2.697	0.024	0.281	2.273	2.659	0.983	0.997	0.986