Continuum Regression and Ridge Regression
本文揭示了第一因子连续回归与标准岭回归的紧密关系，指出连续回归通过引入标量补偿因子，在应对近共线性时原则上优于岭回归，且预测效率相当但对参数选择更不敏感。
SUMMARY 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.