Banca de QUALIFICAÇÃO: TITO LIVIO DA CUNHA LOPES

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
DISCENTE : TITO LIVIO DA CUNHA LOPES
DATA : 25/08/2016
HORA: 14:00
LOCAL: Sala de seminários de estatística CCET
TÍTULO:

Models for integer valued based on the thinning operator


PALAVRAS-CHAVES:

Binary data; Operators. INAR(1) process; Inflated-parameter generalized power series distribution.


PÁGINAS: 81
GRANDE ÁREA: Ciências Exatas e da Terra
ÁREA: Probabilidade e Estatística
SUBÁREA: Estatística
RESUMO:
The first part of the work concerns the binary data modeling autocorrelated in which a
new class of models has been proposed, called Bernoulli process autoregressive order
p - Bear(p) familiar to existing classical models in literature, for example, autoregressive
model - AR(p) and INAR(p). BeAr(1) model has been studied and established
various properties, including three estimation methods were studied for the model.
In addition to simulations, applications were made on real data of precipitation, at
which the BeAr(1) and BeAr(2) models form proposed for modeling. In the second
part, new models were studied to suggest the distribution of family IGPSD for the innovation
process INAR(1) model. The main properties of the process were established,
for exemplo, mean, variance, autocorrelation and transition probability. The estimation
methods for Yule Walker (YW) and conditional maximum likelihood (CML) were applied.
Two particular cases of models INAR(1) with IGPSD innovation process were
studied. In the application to real data overdispersed, the proposed IPoINAR(1) model
had better performed with respect to other models, according to the criterion AIC.

MEMBROS DA BANCA:
Interno - 1217007 - ANDRE LUIS SANTOS DE PINHO
Interno - 308.629.298-90 - FERNANDO FERRAZ DO NASCIMENTO - UFPI
Interno - 2083593 - LUZ MILENA ZEA FERNANDEZ
Presidente - 1023112 - MARCELO BOURGUIGNON PEREIRA
Notícia cadastrada em: 08/08/2016 11:04
SIGAA | Superintendência de Tecnologia da Informação - (84) 3342 2210 | Copyright © 2006-2024 - UFRN - sigaa02-producao.info.ufrn.br.sigaa02-producao