CONTINUUM REGRESSION AND RIDGE-REGRESSION
成果类型：
Article
署名作者：
SUNDBERG, R
刊物名称：
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY
ISSN/ISSBN：
1369-7412
发表日期：
1993
页码：
653-659
关键词：
摘要：
We demonstrate the close relationship between first-factor continuum regression and standard ridge regression. The difference is that continuum regression inserts a scalar compensation factor for that part of the shrinkage in ridge regression that has no connection with tendencies towards collinearity. We interpret this to mean that first-factor continuum regression is preferable in principle to ridge regression if we want protection against near collinearity but do not admit shrinkage as a general principle. Furthermore, our experience indicates that with first-factor continuum regression we can obtain predictors that are at least as mean-squared error efficient as with ridge regression but with less sensitivity to the choice of ridge constant. The scalar compensation factor is easily calculated by just an additional simple linear regression with the ridge regression predictor as regressor.