Condition Numbers and Minimax Ridge Regression Estimators

DOI10.2307/2288496zbMath0575.65148OpenAlexW4239844857MaRDI QIDQ3694648

George Casella

Publication date: 1985

Full work available at URL: https://doi.org/10.2307/2288496

zbMATH Keywords

mean squared errornumerical stabilitymatrix conditioningminimax ridge regression estimator

Mathematics Subject Classification ID

Ridge regression; shrinkage estimators (Lasso) (62J07) Minimax procedures in statistical decision theory (62C20) Probabilistic methods, stochastic differential equations (65C99)

Related Items

More on Liu-Type Estimator in Linear Regression ⋮ LowCon: A Design-based Subsampling Approach in a Misspecified Linear Model ⋮ Bayes minimax ridge regression estimators ⋮ The coefficient of determination in the ridge regression ⋮ Choice of the ridge factor from the correlation matrix determinant ⋮ Fully Bayes factors with a generalized \(g\)-prior ⋮ Robust estimation of common regression coefficients under spherical symmetry ⋮ Enhanced ridge regressions ⋮ Transformation of variables and the condition number in ridge estimation ⋮ A new class of generalized Bayes minimax ridge regression estimators ⋮ Improved minimax estimation of a multivariate normal mean under heteroscedasticity ⋮ Shrinkage and modification techniques in estimation of variance and the related problems: A review