Optimal Bayesian design for model discrimination via classification.

Optimal Bayesian design for model discrimination via classification.
复制标题

DOI:
10.1007/s11222-022-10078-2
复制
发表时间:
2022
影响因子:
2.2
通讯作者:
Drovandi C
Drovandi C
中科院分区:
数学2区
文献类型:
--
作者:
Hainy M;Price DJ;Restif O;Drovandi C

文献摘要

参考文献

相似文献

执行最优贝叶斯设计来区分竞争模型是计算密集型的,因为它涉及估计数千个模拟数据集的后验模型概率。当竞争模型的似然函数在计算上很昂贵时,这个问题会进一步复杂化。提出了一种利用监督分类方法进行贝叶斯最优模型判别设计的新方法。与以前使用近似贝叶斯计算的方法相比,这种方法需要的候选模型模拟要少得多。此外,通过误分类错误率可以很容易地评估优化设计的性能。该方法在具有难以处理的可能性的模型中特别有用,但在可能性可管理的情况下也可以提供计算优势。在线版本包含补充资料,下载地址:10.1007/s11222-022-10078-2。
Performing optimal Bayesian design for discriminating between competing models is computationally intensive as it involves estimating posterior model probabilities for thousands of simulated data sets. This issue is compounded further when the likelihood functions for the rival models are computationally expensive. A new approach using supervised classification methods is developed to perform Bayesian optimal model discrimination design. This approach requires considerably fewer simulations from the candidate models than previous approaches using approximate Bayesian computation. Further, it is easy to assess the performance of the optimal design through the misclassification error rate. The approach is particularly useful in the presence of models with intractable likelihoods but can also provide computational advantages when the likelihoods are manageable. The online version contains supplementary material available at 10.1007/s11222-022-10078-2.
DOI: 10.1214/08-aos635
发表时间: 2009-08-01
影响因子: 4.5
作者:
Dette, Holger;Titoff, Stefanie
通讯作者: Titoff, Stefanie
DOI: 10.1007/s00477-015-1067-8
发表时间: 2016
期刊: Stochastic environmental research and risk assessment : research journal
影响因子: --
作者:
Hainy M;Müller WG;Wagner H
通讯作者: Wagner H
DOI: 10.1016/j.csda.2013.11.003
发表时间: 2014-04-01
影响因子: 1.8
作者:
Drovandi, Christopher C.;Pettitt, Anthony N.;McCombe, Pamela A.
通讯作者: McCombe, Pamela A.
DOI: 10.1002/bimj.201900074
发表时间: 2020-07
影响因子: 1.7
作者:
Harman, Radoslav;Mueller, Werner G.
通讯作者: Mueller, Werner G.
DOI: 10.1080/10618600.2012.730083
发表时间: 2014-03-01
影响因子: 2.4
作者:
Drovandi, Christopher C.;McGree, James M.;Pettitt, Anthony N.
通讯作者: Pettitt, Anthony N.