Σύγκριση μεθόδων

	Ελάχιστα Τετράγωνα (Ordinary Least Squares - OLS)	Παλινδρόμηση Lasso
Πεδίο	Στατιστική	Μηχανική Μάθηση
Οικογένεια	Regression model	Machine learning
Έτος προέλευσης	1805	1996
Δημιουργός	Adrien-Marie Legendre (1805); Carl Friedrich Gauss (1809)	Tibshirani, R.
Τύπος	Linear parameter estimation	Regularized linear regression (L1 penalty)
Θεμελιώδης πηγή	Legendre, A.-M. (1805). Nouvelles méthodes pour la détermination des orbites des comètes. Firmin Didot, Paris. [Appendix: Sur la Méthode des moindres quarrés, pp. 72–80.]	Tibshirani, R. (1996). Regression Shrinkage and Selection via the Lasso. Journal of the Royal Statistical Society: Series B, 58(1), 267–288.
Εναλλακτικές ονομασίες	OLS, OLS regression, linear least squares, classical linear regression	LASSO Regresyonu, lasso, L1-regularized regression, L1 regularization
Συναφείς	8	4
Σύνοψη	Ordinary Least Squares (OLS) is the canonical method for estimating the parameters of a linear regression model by minimizing the sum of squared differences between observed and predicted values. First published by Adrien-Marie Legendre in 1805 and independently developed by Carl Friedrich Gauss (who claimed priority from 1795), OLS is provably optimal under the Gauss-Markov theorem: given its assumptions, it yields the Best Linear Unbiased Estimator (BLUE) of the regression coefficients.	Lasso regression, introduced by Robert Tibshirani in 1996, is a linear regression method that adds an L1 penalty to the loss so that it shrinks coefficients and performs variable selection at the same time, producing a sparse model. By driving some coefficients exactly to zero it keeps only the predictors that matter.