I am reading the book "regression estimators" by Gruber 2010 where he uses this technique to compare Ridge Regressors, however he concentrates on deriving the mathematical results without giving any empirical results/examples.
Question: I am eager to read empirical type papers where these information geometry techniques are used to compare ridge regressors on real data.
Does anyone know of any such papers?
It seems from reading the book that the regressor with the highest variance will have the shorter distance form OLS. So it would seem that the regressor with the biggest distance should be selected but this is not explicitly stated in the book.
I'd also be very interested in hearing about alternate methods to compare ridge regressors and the relative pros/cons of those approaches ideally compared to the informational geometry approach.
Note: A question has already been asked about the usefulness of informational geometry. Using information geometry to define distances and volumes…useful?