Bayesian and non bayesian estimation based on generalized order statistics for lifetime distribution /
التقدير البييزى واللابييزى فى ضوء الإحصاءات الترتيبية المعممة لتوزيع الحياه
Said Gamal Nasser Mohamed Nassr ; Supervised Abdallah Mohamed Abdelfattah , Amal Soliman Hassan
- Cairo : Said Gamal Nasser Mohamed , 2015
- 170 Leaves : charts ; 30cm
Thesis (Ph.D.) - Cairo University - Institute of Statistical Studies & Research - Department of Mathematical Statistics
In this thesis, presents point estimates of parameters for exponentiated Weibull Poisson distribution based on generalized order statistics using maximum likelihood and Bayesian methods. The asymptotic variances and covariance matrix and confidence interval estimates of parameters are derived based on generalized order statistics. Maximum likelihood and Bayesian predictive (point and interval) of the first future observations are obtained based on generalized order statistics. All results are specialized to type II censored data and upper record values. Bayesian prediction estimators of the first future observation from exponentiated Weibull Poisson distribution are obtained by using different loss functions. All estimation algorithms and numerical studies are carried out to asses these effects using MathCAD (14) statistical package
Bayesian method Exponentiated Weibull Poisson distribution Generalized order statistics