https://nova.newcastle.edu.au/vital/access/ /manager/Index ${session.getAttribute("locale")} 5 https://nova.newcastle.edu.au/vital/access/ /manager/Repository/uon:44666 Wed 19 Oct 2022 13:56:20 AEDT ]]> https://nova.newcastle.edu.au/vital/access/ /manager/Repository/uon:14241 Wed 11 Apr 2018 15:32:05 AEST ]]> https://nova.newcastle.edu.au/vital/access/ /manager/Repository/uon:44084 a, …, a) priors with a < 1 have previously been shown to be inadequate in the presence of zero counts, suggesting that the uniform prior (a = 1) is the preferred candidate. In the presence of many zero counts, however, this prior may not be satisfactory either. A model selection approach is proposed, allowing for the possibility of zero parameters corresponding to zero count categories. This approach results in a posterior mixture of Dirichlet distributions and marginal mixtures of beta distributions, which seem to avoid the problems that potentially result from the various proposed Dirichlet priors, in particular in the context of extreme data with zero counts.]]> Thu 06 Oct 2022 15:06:32 AEDT ]]> https://nova.newcastle.edu.au/vital/access/ /manager/Repository/uon:8392 Sat 24 Mar 2018 08:39:41 AEDT ]]> https://nova.newcastle.edu.au/vital/access/ /manager/Repository/uon:8391 Sat 24 Mar 2018 08:39:41 AEDT ]]> https://nova.newcastle.edu.au/vital/access/ /manager/Repository/uon:19981 Sat 24 Mar 2018 07:50:58 AEDT ]]> https://nova.newcastle.edu.au/vital/access/ /manager/Repository/uon:4882 1) seems appropriate, while a beta(a, b) with a < 1 can be too informative. It is thus argued that sensitivity analysis of an informative prior should be based on a consensus posterior corresponding to the Bayes–Laplace prior rather than the Jeffreys prior.]]> Sat 24 Mar 2018 07:22:01 AEDT ]]> https://nova.newcastle.edu.au/vital/access/ /manager/Repository/uon:4748 Sat 24 Mar 2018 07:21:08 AEDT ]]>