Bivariate Log-Symmetric Models: Theoretical Properties and Parametric Estimation
Bivariate Log-symmetric Models. Monte Carlo simulation. maximum likelihood method. R software.
The bivariate Gaussian distribution has been the basis of probability and statistics for many years. Nonetheless, this distribution faces some problems, mainly due to the fact that many real-world phenomena generate data that follow asymmetric distributions. Bidimensional log symmetric models have attractive properties and can be considered as good alternatives to solve this problem. In this dissertation, we propose new characterizations of bivariate log-symmetric distributions and their applications. This dissertation aims to develop important contributions to probability, theoretical and applied statistics due to the flexibility and interesting properties of the outlined models.