IMPROVED ESTIMATION OF REGRESSION PARAMETERS.
STANFORD UNIV CALIF DEPT OF STATISTICS
Pagination or Media Count:
Correspondences between the problems of estimating the mean of a multivariate normal distribution and estimating regression parameters are presented and investigated to obtain minimax or admissible estimators of the regression parameters in normal multivariate and univariate regression models with respect to squared-distance loss functions. These new estimators are better than the maximum likelihood estimator, in that their risks are smaller, for all parameter values. Author
- Statistics and Probability