Particle Learning for Fat-tailed Distributions

dc.contributor.authorHEDIBERT FREITAS LOPES
dc.contributor.authorPolson, Nicholas G.
dc.coverage.cidadeSão Paulopt_BR
dc.coverage.paisBrasilpt_BR
dc.creatorPolson, Nicholas G.
dc.date.accessioned2023-07-25T13:12:50Z
dc.date.available2023-07-25T13:12:50Z
dc.date.issued2014
dc.description.abstractIt is well-known that parameter estimates and forecasts are sensitive to assumptions about the tail behavior of the error distribution. In this paper we develop an approach to sequential inference that also simultaneously estimates the tail of the accompanying error distribution. Our simulation-based approach models errors with a tν-distribution and, as new data arrives, we sequentially compute the marginal posterior distribution of the tail thickness. Our method naturally incorporates fat-tailed error distributions and can be extended to other data features such as stochastic volatility. We show that the sequential Bayes factor provides an optimal test of fat-tails versus normality. We provide an empirical and theoretical analysis of the rate of learning of tail thickness under a default Jeffreys prior. We illustrate our sequential methodology on the British pound/US dollar daily exchange rate data and on data from the 2008-2009 credit crisis using daily S&P500 returns. Our method naturally extends to multivariate and dynamic panel data.
dc.description.otherIt is well-known that parameter estimates and forecasts are sensitive to assump tions about the tail behavior of the error distribution. In this paper we develop an approach to sequential inference that also simultaneously estimates the tail of the accompanying error distribution. Our simulation-based approach models errors with a tν-distribution and, as new data arrives, we sequentially compute the marginal posterior distribution of the tail thickness. Our method naturally incorporates fat-tailed error distributions and can be extended to other data features such as stochastic volatility. We show that the sequential Bayes factor provides an optimal test of fat-tails versus normality. We provide an empir ical and theoretical analysis of the rate of learning of tail thickness under a default Jeffreys prior. We illustrate our sequential methodology on the British pound/US dollar daily exchange rate data and on data from the 2008-2009 credit crisis using daily S&P500 returns. Our method naturally extends to multivariate and dynamic panel data.pt_BR
dc.format.extent39 p.pt_BR
dc.format.mediumDigitalpt_BR
dc.identifier.issueBEWP 203/2014
dc.identifier.urihttps://repositorio.insper.edu.br/handle/11224/5943
dc.language.isoInglêspt_BR
dc.publisherInsperpt_BR
dc.relation.ispartofseriesInsper Working Paperpt_BR
dc.rights.licenseO INSPER E ESTE REPOSITÓRIO NÃO DETÊM OS DIREITOS DE USO E REPRODUÇÃO DOS CONTEÚDOS AQUI REGISTRADOS. É RESPONSABILIDADE DO USUÁRIO VERIFICAR OS USOS PERMITIDOS NA FONTE ORIGINAL, RESPEITANDO-SE OS DIREITOS DE AUTOR OU EDITORpt_BR
dc.subject.keywordsBayesian Inferencept_BR
dc.subject.keywordsMCMCpt_BR
dc.subject.keywordsKullback-Leiblerpt_BR
dc.subject.keywordsDynamic Panel Datapt_BR
dc.subject.keywordsCredit Crisispt_BR
dc.titleParticle Learning for Fat-tailed Distributionspt_BR
dc.typeworking paper
dspace.entity.typePublication
local.subject.cnpqCiências Exatas e da Terrapt_BR
local.typeWorking Paperpt_BR
relation.isAuthorOfPublication41f844cb-0e5a-4ef1-bb19-5ab1cec8e2ca
relation.isAuthorOfPublication.latestForDiscovery41f844cb-0e5a-4ef1-bb19-5ab1cec8e2ca

Arquivos

Pacote original

Agora exibindo 1 - 1 de 1
N/D
Nome:
BEWP_203_2014_Particle_Learning_for_Fat-tailed_Distributions_TC.pdf
Tamanho:
791.53 KB
Formato:
Adobe Portable Document Format