Jeerapa Sappakitkamjorn. Statistical estimation in restricted parameter spaces. Doctoral Degree(Applied Statistics). King Mongkut's University of Technology North Bangkok. Central Library. : King Mongkut's University of Technology North Bangkok, 2013.
Statistical estimation in restricted parameter spaces
Abstract:
This thesis deals with three main issues. The first issue concerns a confidence interval for location parameters, particularly the population mean in restricted parameter space when the distributions of data are normal and a confidence interval for the population mean and the ratio of the means in restricted parameter space when the distributions of data are lognormal. The second issue concerns confidence intervals for the scale parameters, particularly the population variance and the coefficient of variation in restricted parameter space when the distributions of data are normal and lognormal. For these two issues, new confidence intervals are presented and the analytical expressions of coverage probability and expected length are mathematically derived. Finally, the third issue concerns the Jackknife based confidence interval of the difference between two normal population means when data are presented with outliers. A simulation study is conducted to examine the Jackknife based confidence intervals in terms of coverage probability and expected length.