Uma revisão de modelos lineares dinâmicos

Loading...
Thumbnail Image

Journal Title

Journal ISSN

Volume Title

Publisher

Universidade Federal de Juiz de Fora (UFJF)

DOI

Abstract

The state space models are a proposal for time series analysis and forecasting in viewpoint Bayesian. This course conclusion work aims to present the Dynamics Linear Models (DLM), its use and application context. DLM’s properties are shown and demonstrated, like Kalman filter, the smoothing, the forecasting one-step-ahead and so the forecasting steps ahead. Markov chain Monte Carlo methods (MCMC) are reviewed in recursive algorithms used to estimate parameters, they are: Gibbs sampler, MetropolisHastings e FFBS.

Description

Citation

Endorsement

Review

Supplemented By

Referenced By

Creative Commons license

Except where otherwise noted, this item's license is described as Acesso Aberto