A New Efficient Redescending M-Estimator for Robust Fitting of Linear Regression Models in the Presence of Outliers

Table 8

Total mean square error for contaminated datasets generated from (X∼N (0, 1); 80% Y∼N (0, 1) and 20% Y∼N (100, 1)).

N	P	OLS	Huber	Hampel	Andrew	Tukey	Ali
N = 30	P = 2	21.410	0.615	0.314	0.317	0.317	0.307
	P = 3	23.81	0.727	0.398	0.405	0.405	0.390
	P = 6	27.764	3.278	0.608	0.649	0.649	0.599
N = 50	P = 2	20.800	0.559	0.225	0.229	0.229	0.213
	P = 3	21.885	0.681	0.288	0.289	0.289	0.270
	P = 6	24.249	0.858	0.421	0.429	0.428	0.410
N = 100	P = 2	20.467	0.540	0.158	0.160	0.160	0.150
	P = 3	20.883	0.567	0.200	0.205	0.205	0.197
	P = 6	22.150	0.664	0.282	0.288	0.288	0.271