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

Table 9

Total mean square error for datasets generated from X∼N(0, 1); 60% Y∼N(0, 1) and 40% Y∼N(100, 1)).

N	P	OLS	Huber	Hampel	Andrew	Tukey	Ali
N = 30	P = 2	41.297	12.45	0.356	0.356	0.356	0.345
	P = 3	42.357	23.84	0.466	0.467	0.467	0.445
	P = 6	46.583	48.64	0.755	0.752	0.752	0.728
N = 50	P = 2	40.617	5.583	0.260	0.260	0.260	0.245
	P = 3	41.376	14.14	0.337	0.337	0.401	0.327
	P = 6	43.420	34.79	0.497	0.497	0.497	0.477
N = 100	P = 2	40.38	2.297	0.183	0.183	0.183	0.161
	P = 3	40.60	4.247	0.233	0.233	0.269	0.211
	P = 6	42.637	11.335	0.330	0.334	0.334	0.310