Σύγκριση μεθόδων

	Εκτίμηση της Διάμεσης Απόλυτης Απόκλισης (MAD)	Παλινδρόμηση Ελαχίστων Τετραγώνων (OLS)	Παλινδρόμηση Ridge
Πεδίο	Στατιστική	Οικονομετρία	Μηχανική Μάθηση
Οικογένεια	Regression model	Regression model	Machine learning
Έτος προέλευσης	1974	2019	1970
Δημιουργός	Hampel (influence-curve treatment); classical robust statistics	Wooldridge (textbook treatment); classical least squares	Hoerl, A.E. & Kennard, R.W.
Τύπος	Robust scale estimator	Linear regression	L2-regularized linear regression
Θεμελιώδης πηγή	Hampel, F. R. (1974). The Influence Curve and Its Role in Robust Estimation. Journal of the American Statistical Association, 69(346), 383-393.	Wooldridge, J. M. (2019). Introductory Econometrics: A Modern Approach (7th ed.). Cengage Learning. ISBN: 978-1337558860	Hoerl, A.E. & Kennard, R.W. (1970). Ridge Regression: Biased Estimation for Nonorthogonal Problems. Technometrics, 12(1), 55–67.
Εναλλακτικές ονομασίες	median absolute deviation, MAD scale estimator, robust scale estimation, Medyan Mutlak Sapma (MAD) Tahmini	ordinary least squares, classical linear regression, linear regression, en küçük kareler regresyonu	Ridge Regresyonu, ridge regresyonu, L2-regularized regression, Tikhonov regularization
Συναφείς	5	5	4
Σύνοψη	Median Absolute Deviation estimation is a robust measure of statistical dispersion that replaces the standard deviation when outliers are present. Rooted in the influence-curve framework formalised by Hampel (1974), it summarises the spread of a continuous variable using medians instead of means, so a single extreme value cannot distort the result.	Ordinary Least Squares is the classical linear regression method that explains a continuous outcome as a linear combination of predictors. It estimates the coefficients by minimising the sum of squared residuals, and under the Gauss-Markov assumptions these estimates are the best linear unbiased estimator (BLUE).	Ridge Regression is an L2-regularized linear regression method, introduced by Arthur Hoerl and Robert Kennard in 1970, that reduces multicollinearity by adding a penalty on the size of the coefficients. It shrinks coefficients toward zero without setting any of them exactly to zero, producing more stable estimates when predictors are highly correlated.