Return to search

Análise comparativa de estimadores da ordem de cadeias de markov

Dissertação (mestrado)—Universidade de Brasília, Instituto de Ciências Exatas, Departamento de Matemática, 2009. / Submitted by Larissa Ferreira dos Angelos (ferreirangelos@gmail.com) on 2010-04-07T18:50:46Z
No. of bitstreams: 1
2009_PauloAngeloAlvesResende.pdf: 340390 bytes, checksum: af9c1fec2af2d27b2f9d727b0e8c4b19 (MD5) / Approved for entry into archive by Daniel Ribeiro(daniel@bce.unb.br) on 2010-05-13T21:25:23Z (GMT) No. of bitstreams: 1
2009_PauloAngeloAlvesResende.pdf: 340390 bytes, checksum: af9c1fec2af2d27b2f9d727b0e8c4b19 (MD5) / Made available in DSpace on 2010-05-13T21:25:23Z (GMT). No. of bitstreams: 1
2009_PauloAngeloAlvesResende.pdf: 340390 bytes, checksum: af9c1fec2af2d27b2f9d727b0e8c4b19 (MD5)
Previous issue date: 2009 / Neste trabalho estudamos o estimador da ordem de cadeias de Markov usando o critério EDC (Efficient Determination Criterion) com o termo de penalidade ótimo proposto por Dorea (2008). Realizamos uma análise comparativa das performances dos estimadores EDCopt, BIC e AIC, baseada nos resultados de simulações computacionais realizadas. _________________________________________________________________________________ ABSTRACT / In what follows we study and analyze the Markov chain order estimator EDC (Efficient Determination Criterion) with the penalty function proposed by Dorea (2008). We also carry out extensive numerical simulations based on EDC, BIC and AIC, aiming to a detailed comparison of their features as well as their relative performance.

Identiferoai:union.ndltd.org:IBICT/oai:repositorio.unb.br:10482/4611
Date January 2009
CreatorsResende, Paulo Angelo Alves
ContributorsGonçalves, Cátia Regina
Source SetsIBICT Brazilian ETDs
LanguagePortuguese
Detected LanguageEnglish
Typeinfo:eu-repo/semantics/publishedVersion, info:eu-repo/semantics/masterThesis
Sourcereponame:Repositório Institucional da UnB, instname:Universidade de Brasília, instacron:UNB
Rightsinfo:eu-repo/semantics/openAccess

Page generated in 0.002 seconds