Prior Sensitivity Analysis in a Semi-Parametric Integer-Valued Time Series Model

Prior Sensitivity Analysis in a Semi-Parametric Integer-Valued Time Series Model

We examine issues of prior sensitivity in a semi-parametric hierarchical extension of the INAR(<i>p</i>) model with innovation rates clustered according to a Pitman&#8722;Yor process placed at the top of the model hierarchy. Our main finding is a graphical criterion that guides the s...

Full description

Bibliographic Details
Main Authors: Helton Graziadei, Antonio Lijoi, Hedibert F. Lopes, Paulo C. Marques F., Igor Prünster
Format: Article
Language:English
Published: MDPI AG 2020-01-01
Series:Entropy
Subjects:
time series of counts
bayesian hierarchical modeling
bayesian nonparametrics
pitman–yor process
prior sensitivity
clustering
bayesian forecasting
Online Access:https://www.mdpi.com/1099-4300/22/1/69
Description
Summary:We examine issues of prior sensitivity in a semi-parametric hierarchical extension of the INAR(<i>p</i>) model with innovation rates clustered according to a Pitman&#8722;Yor process placed at the top of the model hierarchy. Our main finding is a graphical criterion that guides the specification of the hyperparameters of the Pitman&#8722;Yor process base measure. We show how the discount and concentration parameters interact with the chosen base measure to yield a gain in terms of the robustness of the inferential results. The forecasting performance of the model is exemplified in the analysis of a time series of worldwide earthquake events, for which the new model outperforms the original INAR(<i>p</i>) model.
ISSN:1099-4300

Similar Items