期刊
JOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS
卷 30, 期 4, 页码 958-976出版社
TAYLOR & FRANCIS INC
DOI: 10.1080/10618600.2021.1875839
关键词
Approximate Bayesian computation; Likelihood-free inference; Model misspecification; Robust Bayesian inference; Slice sampling; Synthetic likelihood
资金
- Discovery Early Career Researcher Award funding scheme [DE200101070]
- Australian Research Council [DE200101070] Funding Source: Australian Research Council
Bayesian synthetic likelihood (BSL) is a well-established method for approximate Bayesian inference, but unreliable parameter inference can occur if the assumed data-generating process does not match the actual process. A new approach to BSL has been proposed to detect model misspecification and deliver accurate inferences even when the model is misspecified.
Bayesian synthetic likelihood (BSL) is now an established method for conducting approximate Bayesian inference in models where, due to the intractability of the likelihood function, exact Bayesian approaches are either infeasible or computationally too demanding. Implicit in the application of BSL is the assumption that the data-generating process (DGP) can produce simulated summary statistics that capture the behaviour of the observed summary statistics. We demonstrate that if this compatibility between the actual and assumed DGP is not satisfied, that is, if the model is misspecified, BSL can yield unreliable parameter inference. To circumvent this issue, we propose a new BSL approach that can detect the presence of model misspecification, and simultaneously deliver useful inferences even under significant model misspecification. Two simulated and two real data examples demonstrate the performance of this new approach to BSL, and document its superior accuracy over standard BSL when the assumed model is misspecified. for this article are available online.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据