[ABE-L] Seminário DEST/UFMG em 10/05/2024

[Marcos Prates]

Caros,

Na próxima sexta-feira (10 de Maio, às 13:30h) o ciclo de Seminários do
Departamento de Estatística da UFMG terá a apresentação do prof. Xia Wang
da University of Cincinnati - USA.

Dr. Xia Wang is a professor in the Division of Statistics and Data Science,
Department of Mathematical Sciences at the University of Cincinnati, USA.
Her main research interests focus on Bayesian modeling of categorical data
and scalable modeling of complex, high-dimensional data. Her work completed
in the past few years relates to improve flexibility on joint modeling with
applications to patient data with multiple outcomes collected over time. In
addition, she has
worked on large-scale hypotheses testing on temporally dependent data.
Recently, she got interested in methods and algorithm development in
Bayesian variable selection. Today’s talk is one of the ongoing works with
her PhD student.

Title: Variable Selection for Zero-Inflated Poisson Regression Model

The study implements an efficient algorithm for variable selection in the
zero-inflated count regression model based on Polya-Gamma latent variables.
This leads to a closed form posterior conditional distribution under a
logistic link function in modeling the excessive zeros, which helps
overcome the computational disadvantage of the logistic link compared to a
probit link. Simulation studies examines the efficacy of the proposed model
in selecting important variables
as well as how the choice of link functions, between the two commonly used
probit and the logit links, influences the variable selection results. The
proposed model and its comparison with other methods are also illustrated
through the application to a German Healthcare dataset. This is a joint
work with Haichao Zhang.

O seminário será transmitido ao vivo pelo canal do Youtube "Seminários DEST
- UFMG".

https://www.youtube.com/@seminariosdest-ufmg

Att,
Marcos Prates
-------------- Próxima Parte ----------
Um anexo em HTML foi limpo...
URL: <http://lists.ime.usp.br/pipermail/abe/attachments/20240503/1aa67dbf/attachment.htm>