One of the most important problem in data analysis is data behavior reorganization. So, for this it is necessary to fit a distribution on the data and estimate the parameters of distribution. Off course we need to find the distribution which is more flexible than others for data fitting. It is for the data structure because some data has overdispersion (the variance is greater than the mean) and another one perhaps inflates in a point and so on. In this thesis, we introduce a distribution which is called Uniform-Geometric and we assess some properties of this distribution[1]. For fitting the data recently authors introduce several new distributions like Uniform-Poisson[2], generalized Poisson-Lindley[4] and a new generalization of the geometric distribution[3] that more flexible and appropriate than traditional distributions. They verified several properties of the distributions and show that their distributions can be fitted on several data because of their flexibility.
