![]() | Only 14 pages are availabe for public view |
Abstract The problem of outliers is of important in almost all experimental fields and has received continued attention in mathematical statistics, because it is effect on each distribution, it is effect on means, variances, regression coefficients, bias or distribution of estimates .The literature review of some distributions will be introduced in the presence of one outlier, two outliers and k outliers. Burr distributions will be discussed which were first discussed by Burr (1942) a two parameter family but, I{u2019}ll concern and restrict Burr type III distribution in the presence of outliers. In this thesis, the estimation of the parameters for Burr type III distribution in the presence of k outliers are obtained using moment, maximum likelihood and Bayesian estimation methods |