Research Article
Robust Wild Bootstrap for Stabilizing the Variance of Parameter Estimates in Heteroscedastic Regression Models in the Presence of Outliers
Table 3
Standard errors of the non-robust and robust wild bootstrap.
| % outliers | Coeff. | Bootols | Bootwu | Bootliu | RBootwu | RBootliu |
| Sample Size n = 20 |
| 0% | | | | | |
0.9774
| |
1.2106
|
1.3220
|
1.0294
|
1.2739
|
1.1894
| |
1.6257
|
1.4866
|
1.3845
|
1.4827
|
1.3737
| Mean |
1.3285
|
1.2833
|
1.1226
|
1.2711
|
1.1802
| 5% | | | | | | | |
6.7426
|
8.0992
|
6.5537
|
1.1291
|
1.2546
| |
5.4343
|
4.9812
|
3.2680
|
1.4656
|
1.6396
| Mean |
5.5506
|
5.8501
|
4.3591
|
1.1483
|
1.2912
| 10% | | | | | |
0.9913
| |
12.9386
|
15.0904
|
12.5433
|
1.3296
|
1.15814
| |
15.5132
|
16.8739
|
14.1209
|
1.4812
|
1.1552
| Mean |
13.2472
|
14.8332
|
12.3689
|
1.3200
|
1.1015
|
| Sample Size n = 60 |
| 0% | | | | | |
0.6800
| |
0.6632
|
0.6711
|
0.5644
|
0.7213
| | |
0.9779
|
0.9865
|
0.8358
|
0.9979
|
0.8787
| Mean |
0.7970
|
0.7957
|
0.6828
|
0.8216
|
0.7220
| 5% | | | | | |
0.7043
| |
4.6423
|
5.3209
|
4.6852
|
0.7659
|
0.7000
| |
3.4306
|
2.1242
|
1.7946
|
0.9833
|
0.9184
| Mean |
3.6660
|
3.2777
|
2.8587
|
0.8390
|
0.7742
| 10% | | | | | |
0.7583
| |
6.7028
|
8.5494
|
6.9565
|
0.8880
|
0.8150
| |
8.7297
|
10.9102
|
9.1101
|
1.1144
|
1.0058
| Mean |
7.2612
|
9.1310
|
7.5333
|
0.9396
|
0.8597
|
| Sample Size n = 100 |
| 0% | | | | | | | |
0.6340
|
0.6531
|
0.5712
|
0.6441
|
0.5901
| |
0.6355
|
0.6717
|
0.5055
|
0.5681
|
0.4619
| Mean |
0.5784
|
0.6049
|
0.4848
|
0.5483
|
0.4673
| 5% | | | | | | | |
4.6667
|
4.7285
|
4.7007
|
0.6805
|
0.6310
| |
3.1597
|
1.5921
|
1.3564
|
0.5843
|
0.4947
| Mean |
3.4930
|
2.7632
|
2.6435
|
0.5755
|
0.5048
| 10% | | | | | | | |
5.9250
|
7.4991
|
6.1414
|
0.5324
|
0.4687
| |
7.8683
|
9.8171
|
8.0801
|
0.7999
|
0.7220
| Mean |
6.5041
|
8.1722
|
6.7157
|
0.6107
|
0.5429
|
|
|