[ABE-L] Seminar STODAD - Marcio Laurini hoje às 16:30 - Generalized long memory models

[Airlane Pereira Alencar]

Prezados
Hoje teremos nosso seminário às 16:30 com a apresentação do prof. Marcio
Laurini.
Google Meet Link da videochamada: https://meet.google.com/mfz-rgva-yzw
<Seminário STODAD - Marcio Laurini Terça-feira, 31 de maio · 4:30 até
5:30pm Informações de participação do Google Meet Link da videochamada:
https://meet.google.com/mfz-rgva-yzw>
*ou no site*

*https://stream.meet.google.com/stream/4b5670d9-9fab-437c-858c-74a9494e77da
<https://stream.meet.google.com/stream/4b5670d9-9fab-437c-858c-74a9494e77da>*

*Estimating generalized long memory models using integrated nested Laplace
approximations*

*Márcio Laurini (joint work with Pedro Chaim and Fernanda Valente)*

*FEARP-USP*

*In this work we discuss the use of Markovian approximations and the use of
integrated nested Laplace approximations in the estimation of multifactor
dynamic models with long memory. The evolution of the latent factors is
given by fractional Gaussian noise processes, which are approximated with a
weighted sum of independent first-order autoregressive components. This
structure can be represented as a Gaussian Markov Random Field, allowing
the use of computationally efficient Bayesian methods. These approximations
are used in the estimation of multifactor and multivariate stochastic
volatility models, in the estimation of multifactor models for the term
structure of interest rates and in multifactor models for river flow.*


Até mais

Airlane P. Alencar

Professora Associada

Departamento de Estatística


Instituto de Matemática e Estatística

Universidade de São Paulo

Rua do Matão, 1010 - CEP 05508-090 - São Paulo, SP

www.ime.usp.br
-------------- Pr�xima Parte ----------
Um anexo em HTML foi limpo...
URL: <http://lists.ime.usp.br/pipermail/abe/attachments/20220531/17b091a2/attachment.htm>