Non-parametric bayesian inference applied on estimation and forecasting of time series

• Grant number: 12/03000-9
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: July 01, 2012
• End date: December 31, 2012
• Field of knowledge: Physical Sciences and Mathematics - Probability and Statistics - Statistics
• Principal Investigator: Fernando Antonio Moala
• Grantee: Marcelo Hartmann
• Host Institution: Faculdade de Ciências e Tecnologia (FCT). Universidade Estadual Paulista (UNESP). Campus de Presidente Prudente. Presidente Prudente , SP, Brazil

Abstract

Gaussian processes jointly with Bayesian methods are applied to the estimation of functions, that’s it, there is no supposition of a unique parametric functional form underlying the data. In this approach of non-parametric inference, the density function or the function of interest can be estimated without the imposition of any restrictive supposition of its form. The data allow determining the estimate of the interested function rather than conditioning it to given parametric class functions. It also performed some applications in time series models.(AU)