Abstract
Abstract Müller et al. (Stat Methods Appl, 2017) provide an excellent review of
several classes of Bayesian nonparametric models which have found widespread application in a variety of contexts, successfully highlighting their flexibility in comparison
with parametric families. Particular attention in the paper is dedicated to modelling
spatial dependence. Here we contribute by concisely discussing general computational
challenges which arise with posterior inference with Bayesian nonparametric models
and certain aspects of modelling temporal dependence.
Lingua originale | English |
---|---|
pagine (da-a) | 231-238 |
Numero di pagine | 8 |
Rivista | STATISTICAL METHODS & APPLICATIONS |
Volume | 27 |
DOI | |
Stato di pubblicazione | Pubblicato - 2018 |
Keywords
- Bayesian dependent model
- Computation
- Conjugacy
- Dirichlet
- Probability and Uncertainty
- Statistics
- Statistics and Probability
- Transition function