Compara mètodes

	Regressió lineal simple robusta	Mínims Quadrats Ponderats (WLS)
Camp	Estadística	Estadística
Família	Regression model	Regression model
Any d'origen	1964-1987	1935
Autor original	Peter J. Huber (M-estimators, 1964); Rousseeuw & Leroy (practical framework, 1987)	Alexander Craig Aitken
Tipus	Robust linear regression	Weighted linear estimator
Font seminal	Rousseeuw, P. J., & Leroy, A. M. (1987). Robust Regression and Outlier Detection. John Wiley & Sons. ISBN: 978-0471852339	Aitken, A. C. (1935). IV.—On least squares and linear combination of observations. Proceedings of the Royal Society of Edinburgh, 55, 42–48.
Àlies	robust SLR, M-estimator simple regression, outlier-resistant simple regression, robust bivariate regression	WLS, weighted regression, heteroscedasticity-corrected OLS, variance-weighted least squares
Relacionats	6	3
Resum	Robust simple linear regression fits a straight line through bivariate data using loss functions or weighting schemes that down-weight outliers, producing slope and intercept estimates that are far less sensitive to extreme observations than ordinary least squares while remaining easy to interpret.	Weighted Least Squares is a generalization of Ordinary Least Squares (OLS) regression that assigns each observation a weight inversely proportional to its error variance, thereby down-weighting high-variance data points and up-weighting precise ones. Introduced in its general matrix form by Alexander Craig Aitken in 1935, WLS is the canonical remedy when heteroscedasticity is present and the error variance structure is known or can be reliably estimated.