2015
DOI: 10.1080/23737484.2016.1190307
|View full text |Cite
|
Sign up to set email alerts
|

Bayesian GARMA models for count data

Abstract: Generalized autoregressive moving average (GARMA) models are a class of models that was developed for extending the univariate Gaussian ARMA time series model to a flexible observation-driven model for non-Gaussian time series data. This work presents a Bayesian approach for GARMA models with Poisson, binomial, and negative binomial distributions. A simulation study was carried out to investigate the performance of Bayesian estimation and Bayesian model selection criteria. In addition, three real data sets wer… Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
4
1

Citation Types

0
9
0

Year Published

2018
2018
2023
2023

Publication Types

Select...
6
1

Relationship

0
7

Authors

Journals

citations
Cited by 9 publications
(9 citation statements)
references
References 21 publications
0
9
0
Order By: Relevance
“…There are several forms of the Negative binomial in the literature. Andrade et al (2015), for example, considered the form where ν = 1/σ is supposed to be known, belonging to the exponential family and resulting in a GARMA(p, q) submodel.…”
Section: Negative Binomialmentioning
confidence: 99%
See 2 more Smart Citations
“…There are several forms of the Negative binomial in the literature. Andrade et al (2015), for example, considered the form where ν = 1/σ is supposed to be known, belonging to the exponential family and resulting in a GARMA(p, q) submodel.…”
Section: Negative Binomialmentioning
confidence: 99%
“…A Bayesian analysis of the GARMA(p, q) class was presented and discussed in the study of Andrade, Andrade, and Ehlers (2015), who used conditional distributions such as Poisson, Binomial, and Negative binomial. The authors indicated contributions using this approach in terms of point estimation and the range of the credible intervals of the parameters when modeling count data.…”
Section: Introductionmentioning
confidence: 99%
See 1 more Smart Citation
“…On the other hand, a different stream explored integervalued time series counts such as ARMA structures as in [6,5] or INGARCH structure as done in [51,54,52,53]. However, from a Bayesian perspective, the only work to the best of our knowledge is that of [45] where the authors discussed an ARMA model for different count series parameters. However, their treatment of ignoring zero-valued data or putting the MA structure by demeaned Poisson random variable remains questionable.…”
Section: Introductionmentioning
confidence: 99%
“…On the other hand, a different stream explored integer-valued time series counts such as ARMA structures as in (Brandt and Williams, 2001;Biswas and Song, 2009) or INGARCH structure as done in Zhu (2011Zhu ( , 2012c. However, from a Bayesian perspective, the only work to the best of our knowledge is that of Silveira de Andrade et al (2015) where the authors discussed an ARMA model for different count series parameters. However, their treatment of ignoring zero-valued data or putting the MA structure by demeaned Poisson random variable remains questionable.…”
Section: Introductionmentioning
confidence: 99%