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

Table 2

Total mean square error for datasets generated from t-distribution (X & Y∼t-distribution, t(3)).

N	P	OLS	Huber	Hampel	Andrew	Tukey	Ali
N = 30	P = 2	0.373	0.301	0.311	0.305	0.303	0.300
	P = 3	0.441	0.361	0.370	0.380	0.378	0.301
	P = 6	0.671	0.507	0.532	0.660	0.652	0.530
N = 50	P = 2	0.288	0.210	0.221	0.214	0.217	0.240
	P = 3	0.342	0.260	0.266	0.261	0.263	0.267
	P = 6	0.458	0.360	0.369	0.380	0.370	0.384
N = 100	P = 2	0.199	0.157	0.163	0.155	0.155	0.162
	P = 3	0.227	0.172	0.175	0.173	0.175	0.180
	P = 6	0.300	0.237	0.245	0.243	0.243	0.245