Bayes Estimates of the Variance of a Normal Population for Prior Conjugate Distributions of Independent Parameters with application to Estimation in Finite Populations.
CASE WESTERN RESERVE UNIV CLEVELAND OHIO DEPT OF MATHEMATICS AND STATISTICS
Pagination or Media Count:
A Byes estimator of the variance of a normal distribution nmu, sigma-squared, when mu is unknown, is developed for squared-error loss and conjugate priors of independent parameters. In the present study its formula is developed and its relative efficiency is compared with that of the Bayes estimator and with that of the best equivariant estimator. Application of the estimator to the estimation of the variance of a finite population is provided.
- Statistics and Probability