Abstract :
When two distributions have ,approximately ,the same characteristics, it is often difficult to discriminate between them. In this study, we use the ratio of likelihoods for selecting between the logistic and Gumbel distributions for describing a set of data. The parameters for the logistic and Gumbel distributions are estimated by using maximum likelihood (ML), moments (MOM) and order statistic (OS) methods. In addition, by using Monte Carlo simulations, discriminating between the two distributions is investigated in terms of the probability of correct selection (PCS) as found based on the different methods of estimation. In general, it is found that the method of ML outperforms all the other methods when the estimators considered are compared in term of efficiency