Thesis (Ph.D.) - Cairo University - Institute of Statistical Studies and Research - Department of Statistics and Econometrics

The statisticians are interested in constructing flexible distributions to facilitate modeling of lifetime data. So they made generalization of some lifetime models. Although many distributions are discussed in the literature, few regression models have been proposed. In this thesis, a new regression failure time model is introduced, which depend on a location-scale approach. The proposed model is called the log-beta log-logistic regression model. The maximum likelihood estimates of the model parameters for censored data are obtained. The appropriate matrices for assessing the local influence on the parameter estimates under different perturbation schemes are derived. Also the sensitive analysis is measured through selected methods. The performance of the proposed model is testing using both its mean square error and relative root mean square error through simulated data sets. The proposed model is also applied to real medical data sets. Finally, the empirical distribution of the martingale residuals, for the proposed model, is displayed against the normal distribution for comparison. These studies suggest that the martingale residual has normal shape.hesis

Issued also as CD