Multicollinearity and the value of a priori information |
| |
| Authors: | Thomas E. Fomby R. Carter Hill |
| |
| Affiliation: | 1. Department of Economics , Southern Methodist University ,;2. University of Georgia , |
| |
| Abstract: | A measure of multicollinearity is defined which is useful in evaluating maintained hypotheses and aiding estimator selection as it suggests when a non-traditional estimator proposed by Bock (1975) is minimax and dominates ordinary least squares. An example is used to illustrate the presented methodology. |
| |
| Keywords: | principal components regression restricted least squares stein-rule estimators |
|
