Sala de seminários do Departamento de Estatística- UFSCar-São
Carlos-SP
Titulo: Robust
Multivariate Measurement Error Models
with Scale Mixtures of Skew-Normal
Distribution
Autor: Víctor Hugo Lachos Dávila
Departamento de
Estatística, Universidade Estadual de Campinas
Abstract:
Scale
mixtures of skew--normal distribution is a class of
asymmetric
thick--tailed distributions that includes the skew--normal
distribution as
a especial case. The main virtue of some members of this
class of
distributions is that they are easy to simulate from and they make
it
possible to implement the Gibbs Sampler and the EM algorithm
for
parameters estimation. In this paper, we take the scale mixtures of
skew--normal distribution for the unobserved value of the covariates
and
symmetric scale mixtures of normal distribution for the random
errors
providing an appealing robust alternative to the usual symmetric
process
in multivariate measurement error models. Specific distributions
examined
include univariate and multivariate versions of the skew--normal,
the
skew--t, the skew--slash and the skew--contaminated normal
distribution.
The results and methods are applied to a real data
set.
Estão todos convidados, abraços
Josemar
Para mais detalhes
consultar a página: www.ufscar.br/~des