• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 1
  • Tagged with
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
1

Uma abordagem clássica e bayesiana para os modelos de Gompertz e de Richards heteroscedásticos.

Buzolin, Prescila Glaucia Christianini 16 September 2005 (has links)
Made available in DSpace on 2016-06-02T20:06:11Z (GMT). No. of bitstreams: 1 DissPGCB.pdf: 1168050 bytes, checksum: 6dc9351b4fed81fa76650df3ca9d8772 (MD5) Previous issue date: 2005-09-16 / This work presents a classical and a Bayesian approaches to two sigmoidal grownth curves, the Gompertz and the Richards models. We consider the homoscedastic assumption and a multiplicative heteroscedastic structure. For the classical approach we use the maximum likelihood method and for bayesian approach we consider non-informative priors. The posterioris summaries were obtained by the use of the Metropolis-Hastings algorithm. The illustration of both approaches is made using a simulated and a real data set. / Esta dissertação apresenta as abordagens Clássica e Bayesiana para os modelos de crescimento sigmoidais de Gompertz e de Richards. São consideradas as suposições de homoscedasticidade e heteroscedasticidade multiplicativa dos erros. Para a análise Clássica foi utilizado o método de máxima verossimilhança onde a obtenção das estimativas dos parâmetros ocorreu através de métodos iterativos. Para a análise bayesiana, foram consideradas prioris não informativas de Jeffreys e para a obtenção dos resumos a posteriori utilizamos o algoritmo de Metropolis-Hastings. Ambos os métodos foram ilustrados através de dados simulados e reais.

Page generated in 0.1118 seconds