Expand description
Robust statistical methods for StatOxide
This module implements robust regression and estimation methods that are less sensitive to outliers and violations of classical assumptions.
§Methods Implemented
- M-estimators: Huber, Tukey’s biweight, Hampel, Andrews
- S-estimators: High breakdown point estimators
- MM-estimators: Combine high breakdown and high efficiency
- LTS/LMS: Least Trimmed Squares / Least Median of Squares
- Robust covariance estimation: Minimum Covariance Determinant (MCD)
Structs§
- Least
Trimmed Squares - Least Trimmed Squares estimator (high breakdown)
- MEstimator
- M-estimator for robust regression
- MMEstimator
- MM-estimator (combines high breakdown and high efficiency)
- Robust
Regression Results - Robust regression results
- SEstimator
- S-estimator (high breakdown point)
- Tuning
Parameters - Tuning parameters for robust estimators
Enums§
- Loss
Function - Loss functions for M-estimation
- Scale
Estimator - Scale estimation methods