A condition for best asymptotic normality in a regular family of distributions |
| |
Authors: | RP Kindermann MS Levy |
| |
Institution: | Department of Mathematics and Statistics , University of Nebraska - Lincoln , Lincoln, Nebraska, 68588 |
| |
Abstract: | A theorem is presented which provides a simple sufficient condition for a weakly consistent estimator of a parameter in a regular family of distributions to be best asymptotically normal (B.A.N.). As a corollary the B.A.N. property of a maximum likelihood estimator is established under weaker conditions than those of Zacks (1971). Two examples are provided to illustrate the technique. |
| |
Keywords: | best asymptotically normal regular family consistency maximum likelihood estimation |
|
|