Improving bias-robustness of regression estimates through projections
We define a robust procedure to "correct" a regression estimate β̂0 along the directions in predictor space where the fit is worse. When β̂0 is the least median of squares estimate, the "corrected estimate" has a smaller maximum asymptotic bias under contamination, and a much bet...
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | JOUR |
| Acceso en línea: | http://hdl.handle.net/20.500.12110/paper_01677152_v47_n2_p149_Maronna |
| Aporte de: |
| Sumario: | We define a robust procedure to "correct" a regression estimate β̂0 along the directions in predictor space where the fit is worse. When β̂0 is the least median of squares estimate, the "corrected estimate" has a smaller maximum asymptotic bias under contamination, and a much better finite-sample behavior than β̂0. © 2000 Elsevier Science B.V. |
|---|