A class of minimax estimators of a normal quantile |
| |
| Authors: | Andrew L Rukhin |
| |
| Affiliation: | Statistics Department, Purdue University, West Lafayette, IN 47907, U.S.A. |
| |
| Abstract: | The estimation problem of the quantiles of a normal distribution with both parameters unknown, is considered. We construct a class of minimax procedures each of which improves upon the traditional (best equivariant) estimator of a quantile different from the median. For this purpose a differential inequality and a family of its solutions is found. |
| |
| Keywords: | Normal quantiles estimation quadratic loss minimaxness equivariant procedures admissibility differential inequality |
| 本文献已被 ScienceDirect 等数据库收录！ |
