作者: Diane Galarneau Gibbons
DOI: 10.1080/01621459.1981.10477619
关键词:
摘要: Abstract Ridge regression, which defines a class of estimators indexed by biasing parameter k, is an alternative to least squares estimation in the multiple linear regression model. Many algorithms for have been proposed statistical literature. The present study identifies 10 promising and systematically evaluates compares them using Monte Carlo methods. Three perform well overall. A two-parameter estimator offers potential improvement over one-parameter ridge estimators.