Comparação das distribuições α-estável, normal, t de student e Laplace assimétricas

• Autor(a) principal: Macerau, Walkiria Maria de Oliveira
• Data de Publicação: 2012
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da UFSCAR
• Texto Completo: https://repositorio.ufscar.br/handle/ufscar/4555
• topic: Estatística; Distribuição normal assimétrica; Distribuição t-assimétrica; Distribuições assimétricas; t de Student assimétrica; Laplace assimétrica; Asymmetric distributions; Skew normal; Skew t de student; Skew laplace; CIENCIAS EXATAS E DA TERRA::PROBABILIDADE E ESTATISTICA::ESTATISTICA

Abstract The asymmetric distributions has experienced great development in recent times. They are used in modeling financial data, medical, genetics and other applications. Among these distributions, the Skew normal (Azzalini, 1985) has received more attention from researchers (Genton et al., (2001), Gupta et al., (2004) and Arellano-Valle et al., (2005)). We present a comparative study of _-stable distributions, Skew normal, Skew t de Student and Skew Laplace. The _-stable distribution is studied by Nolan (2009) and proposed by Gonzalez et al., (2009) in the context of genetic data. For some real datasets, in areas such as financial, genetics and commodities, we test which distribution best fits the data. We compare these distributions using the model selection criteria AIC and BIC.