Answers are grounded in this paper and cite their sources.
The dependence of ridge-regression estimates on the choice of parameterization is defined with reference to the hull of all possible ridge estimates for all possible parameterizations. A reanalysis of Hoerl and Kennard's (1970) 10-factor example and Marquardt and Snee's (1975) acetylene data reveals that most conclusions depend on the parameterization.
No takes yet. Share an insight, caveat, or question.
Edward E. Leamer (1981) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: