Robust statistical procedures

摘要: Background - Why Robust Procedures? Qualitative and Quantitative Robustness, Breakdown, Infinitesimal Influence Function M-L R-Estimates, L-Estimates, Asymptotic Properties of M-Estimates, Asymptotically Efficient M-L, Scaling Question Minimax Theory, Bias, Variance Multiparameter Problems, Generalities, Regression, Covariances the Affinely Invariant Case, Coordinate Dependent Case Finite Sample Tests Capacities, Estimation Adaptive Estimates, Estimates Robustness Where Are We Now?, The First Ten Years Functions Pseudovalues, Breakdown Outlier Detection, Studentizing, Shrinking Neighbourhoods, Some Persistent Misunderstandings, Future Directions.