Research Article

Robust Wild Bootstrap for Stabilizing the Variance of Parameter Estimates in Heteroscedastic Regression Models in the Presence of Outliers

Table 4

Robustness measure of RMSE of the non-robust and robust wild bootstrap.

Robustness measure of RMSE
% outliersCoeff.BootolsBootwuBootliuRBootwuRBootliu

Sample Size n = 20

0% 110.3099
94.37567 117.3951 94.5929 109.1911
109.20311 117.8290 118.0043
Mean 104.6296 118.5594 99.25083 114.7917
5%
113.8576
94.03987 109.5074
Mean 24.52087 24.25439 32.48111 102.3267 119.4623
10% 9.0949 8.4203
9.1324 8.0386 9.3035 98.6716
9.6977 9.1578 108.3857 109.2154
Mean 9.3083 8.5389 9.8642 101.7699 120.5487

Sample Size n = 60

0%
117.4509 109.0760
99.10685 116.9525 111.2705
Mean 100.1318 116.6189 96.73103 110.2281
5%
14.22979 12.43393 14.13294 80.01583 93.96979
28.49122 45.93069 54.47049 90.03794 106.4720
Mean 22.78076 29.90925 34.79292 87.64955 102.3232
10% 9.4115
9.7919 7.7093 9.4474 74.4197 81.2251
11.1437 10.6705 87.7518 97.1719
Mean 10.89023 8.6814 10.50947 84.6905 92.41533

Sample Size n = 100

0%
97.26561 110.5779 97.95562 108.6698
94.65821 125.5736 137.4324
Mean 95.66548 119.7817 99.45744 125.9034
5% 17.5633
13.5688 13.6576 87.9910 98.4320
20.1261 39.8355 46.7975 90.0007 128.7815
Mean 17.1542 25.6774 28.4363 89.1620 115.7234
10%
Mean 8.9278 7.1393 8.6333 86.2791 95.2121