Research Article
A New Efficient Redescending M-Estimator for Robust Fitting of Linear Regression Models in the Presence of Outliers
Table 3
Total mean square error for datasets generated from Laplace distribution (X & Y∼L(0, 1)).
| N | P | OLS | Huber | Hampel | Andrew | Tukey | Ali |
| N = 30 | P = 2 | 0.336 | 0.293 | 0.305 | 0.300 | 0.299 | 0.227 | P = 3 | 0.415 | 0.349 | 0.365 | 0.372 | 0.370 | 0.305 | P = 6 | 0.562 | 0.507 | 0.532 | 0.685 | 0.675 | 0.529 |
| N = 50 | P = 2 | 0.243 | 0.214 | 0.225 | 0.221 | 0.220 | 0.201 | P = 3 | 0.303 | 0.259 | 0.270 | 0.261 | 0.261 | 0.201 | P = 6 | 0.562 | 0.505 | 0.532 | 0.685 | 0.673 | 0.528 |
| N = 100 | P = 2 | 0.172 | 0.149 | 0.157 | 0.150 | 0.150 | 0.102 | P = 3 | 0.200 | 0.172 | 0.181 | 0.172 | 0.173 | 0.165 | P = 6 | 0.275 | 0.235 | 0.245 | 0.235 | 0.236 | 0.216 |
|
|