![]() | Only 14 pages are availabe for public view |
Abstract In this thesis, we introduce new generalized distributions to the slash distribution. We Have shown that the resulting distributions had the advantage of heavy-tailed distributions for the previous slash. In Chapter 1, we give the definition of heavy tails in distributions and historical and scientific background. Also, we have provided some basic information about the probability distributions that have been used in this thesis. In Chapter 2, we introduce a new generalized family of slash distribution and derive its probability density function for univariate and multivariate forms. We call it double slash distribution. We show that the multivariate double slash distribution is invariant under linear transformation. Furthermore, the moments and marginal distributions are discussed. The maximum likelihood estimation of parameters is also studied. A simulation study is performed to investigate asymptotically the bias properties of the estimators. In Chapter 3, we introduce a new generalized family of slash student distribution and derive its probability density function for univariate and multivariate forms. We call it generalized slash student distribution. We prove that the multivariate generalized slash student distribution is invariant under linear transformation. Furthermore, the moments and marginal distributions are discussed. The maximum likelihood estimation of parameters is also studied. A simulation study is performed to investigate asymptotically the bias properties of the estimators. In Chapter 4, we introduce a new positive distribution and derive its probability density function in univariate form. We call it the log-slash distribution. The basic properties of the log-slash distribution with full proofs are presented. Furthermore, its moments are discussed. The maximum likelihood estimation of parameters is also studied. |