الفهرس Only 14 pages are availabe for public view
 |  | from | 146 |  |  |
| | | from | 146 | | |
Abstract
The problem of estimating the parameters of a probability distribution from a sample is crucial to many fields of science and engineering, particularly for predicting future behaviour of a phenomenon from previously observed sample data. This thesis aims to estimate the unknown parameters of the exponentiated Burr type III distribution from both complete and censored samples. Based on complete sample, the unknown parameters of exponentiated Burr type III distributions are derived under two cases. Assuming that, the three parameters are unknown in first case. While in the second case, one parameter is known and the other two parameters are unknown. Five different methods of estimation are applied, namely, maximum likelihood, moments, percentiles, least squares and weighted least squares. The biases and MSEs are calculated to compare different estimators for all different estimation methods, via simulation using MathCAD (14). Based on type I and type II censored samples, the maximum likelihood method of estimation is used to obtain point estimators for the unknown parameters. Fisher information matrix and confidence intervals are given in case of maximum likelihood estimators, using complete and censoring samples of type I and type II