الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 155 |  |  |
| | | from | 155 | | |
Abstract
Generalized distributions are widely used in various areas of applied fields, such as engineering, medicine, insurance, and others. The main objective of this thesis is to introduce a new generalized classes of distributions. The new distributions contain well-known distributions as special cases and can better fit the real data sets than related distributions. This thesis consists of four chapters presented as follows. Chapter I gives a survey on the historical background of generalized classes of distributions. Some basic definitions, tests of goodness of fit, and measures for best model selection are introduced. A modified version of the half-logistic distribution is derived in Chapter II, called flexible half-logistic distribution. The statistical properties of this new version are studied. Different estimation methods are used to calculate the proposed model estimator. A real data set is used to show the flexibility of the new version. Also, we derive the discrete distribution of the new version and study it. In Chapter III, an extension of the exponentiated log-logistic distribution is derived, called alpha power exponentiated log-logistic distribution. It is represented by its probability density function and cumulative distribution function. Some of its statistical properties are calculated, and its estimated parameters are determined. Finally, we use fuzzy data set to show the flexibility of the new model. I Chapter IV aims to derive a new method for generalized discrete distributions. A generalization of geometric distribution is derived, the statistical properties of the new model are studied, and we determine the estimators of the parameters of this model. It is shown that the proposed model is more flexible than other related models. Also, we generalize the discrete flexible half-logistic distribution.