Ridge regression
Gary C. McDonald · Wiley Interdisciplinary Reviews Computational Statistics · 2009
Abstract Ridge regression is a popular parameter estimation method used to address the collinearity problem frequently arising in multiple linear regression. The formulation of the ridge methodology is reviewed and properties of the ridge estimates capsulated. In particular, four rationales leading to a regression estimator of the ridge form are summarized. Algebraic properties of the ridge regression coefficients are given, which elucidate the behavior of a ridge trace for small values of the ridge parameter (i.e., close to the least squares solution) and for large values of the ridge parameter. Further properties involving coefficient sign changes and rates‐of‐change, as functions of the ridge parameter, are given for specific correlation structures among the independent variables. These results help relate the visual behavior of a ridge trace to the underlying structure of the data. Copyright © 2009 John Wiley & Sons, Inc. This article is categorized under: Statistical Models > Linear Models Algorithms and Computational Methods > Least Squares
Cited by 823 works. doi:10.1002/wics.14
Authors
References
- Regression Shrinkage and Selection Via the Lasso
- The Elements of Statistical Learning
- Ridge Regression: Biased Estimation for Nonorthogonal Problems
- Ridge Regression: Applications to Nonorthogonal Problems
- Generalized Inverses, Ridge Regression, Biased Linear Estimation, and Nonlinear Estimation
- A Biometrics Invited Paper. The Analysis and Selection of Variables in Linear Regression
- Application of ridge analysis to regression problems
- Ridge regression:some simulations
- A Critique of Some Ridge Regression Methods
- A Monte Carlo Evaluation of Some Ridge-Type Estimators