Não foi possível enviar o arquivo. Será algum problema com as permissões?
Diferenças
Aqui você vê as diferenças entre duas revisões dessa página.
| Ambos lados da revisão anterior Revisão anterior Próxima revisão | Revisão anterior | ||
|
projetos:ehlers:par [2007/09/07 11:17] ehlers |
projetos:ehlers:par [2009/03/19 15:14] (atual) ehlers |
||
|---|---|---|---|
| Linha 1: | Linha 1: | ||
| - | === Resumo === | + | === Abstract === |
| In this project we study problems of inference and forecasting in autoregressive models with periodic correlation from a Bayesian perspective. Normality and unimodality assumptions are rarely verified in practice and the usual approach is to try Box-Cox transformations to obtain approximate normality and stabilize the periodic variance. More recently, mixture models were developed to take into account asymmetry and multimodality. | In this project we study problems of inference and forecasting in autoregressive models with periodic correlation from a Bayesian perspective. Normality and unimodality assumptions are rarely verified in practice and the usual approach is to try Box-Cox transformations to obtain approximate normality and stabilize the periodic variance. More recently, mixture models were developed to take into account asymmetry and multimodality. | ||
| === Participants === | === Participants === | ||
| - | - [[pessoais:ehlers|Ricardo Ehlers]] (UFPR) | + | - [[pessoais:ehlers|Ricardo Ehlers]] |
| - | - Marinho Gomes de Andrade (USP) | + | - Marinho Gomes de Andrade |
| === Some references === | === Some references === | ||