Bootstrap for the normal parameters

Abstract

This paper deals with the bootstrap method for the normal parameters. We apply this method to the sample median for estimating the location parameter, and the standard deviation and range for estimating the scale parameter.The accuracy of these estimators with and without outliers are examined through Monte Carlo experiments. They are compared with the usual estimators likethe sample mean and the best linear unbiased estimator. The results of this study show that the proposed estimators have a high accuracy and they are robust against outliers.

Keywords:

• bootstrap
• normal parameters
• order statistics
• robust estimation
• outliers