On estimation problems for competing risks /
عن مشاكل التقدير للأخطار المتنافسه
Wael Shehta Abuelazm Mansour ; Supervised Samir Kamel Ashour
- Cairo : Wael Shehta AbuElazm Mansour , 2016
- 149 Leaves ; 30cm
Thesis (Ph.D.) - Cairo University - Institute of Statistical Studies and Research - Department of Mathematical Statistics
When the life times of items are relatively large or the sample size is large, it is often necessary to terminate the test before all observations are failed, in this case the results will be happening in a censored test. In this thesis we analyze the life time data in the case of two causes of failures (competing risks) based on six censoring scheme namely, Type-I , Type-II, Type-I hybrid, Type-II progressive , Type-I hybrid progressive and Type-II hybrid progressive censoring schemes. We investigate the maximum likelihood and bayesian estimation methods to obtain the estimates of the unknown parameters for some life time distributions. The bayes estimates of the unknown parameters are obtained based on squared error and LINEX loss functions. The new results are reported in chapter IV and V. in chapter IV, we introduce new scheme called progressive hybrid Type-II censoring scheme in the presence of competing risks (Type-II PHCS). Based on this scheme and assumed that the lifetimes of the failure times have an exponential distribution, the maximum likelihood and bayes estimators of the distribution parameters are obtained, and the asymptotic confidence intervals and bayes credible intervals are also proposed. In chapter V, kundu and joarder (2006) present type-I hybrid progressive censoring scheme (Type-I PHCS) in the presence of competing risks when the cause of failure of each item is known. We considered the case when the competing risks have generalized weibull distributions