الفهرس | Only 14 pages are availabe for public view |
Abstract The twenty-first century begun with extending new methods for modern statistics, one of the important developments is to define new useful models through propose new generating families. These models are tested on real data sets available from simple to complex different phenomenon.The main idea of this thesis is to introduce two new bounded generated families of distributions based on a new transformation which presented for the first time.The new transformation considers more flexible to generate new families of continuous distributions with better results for goodness of fit tests.The first family, so called alternative Kumaraswamy-generated family, is proposed by taking the Kumaraswamy distribution as a generator.The second family, called new Topp Leone-generated family is obtained by taking the Topp Leone distribution as a generator. For each family, some new distributions and some statistical properties are derived. Furthermore, the maximum likelihood estimators and a numerical study are discussed for one sub model from each family.Finally, practical data sets are analyzed to illustrate the importance and flexibility of one particular distribution from each family compared with other known models |