A Novel Method of Marginalisation using Low Discrepancy Sequences for Integrated Nested Laplace Approximations
Name:
Articlefile1.pdf
Size:
1.726Mb
Format:
PDF
Description:
Post-print
Embargo End Date:
2022-11-12
Type
ArticleKAUST Department
Statistics ProgramComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
Date
2020-11-21Preprint Posting Date
2019-11-22Online Publication Date
2020-11-21Print Publication Date
2021-05Embargo End Date
2022-11-12Submitted Date
2019-11-22Permanent link to this record
http://hdl.handle.net/10754/665955
Metadata
Show full item recordAbstract
Recently, it has been shown that approximations to marginal posterior distributions obtained using a low discrepancy sequence (LDS) can outperform standard grid-based methods with respect to both accuracy and computational efficiency. This recent method, which we will refer to as LDS-StM, can also produce good approximations to multimodal posteriors. However, implementation of LDS-StM into integrated nested Laplace approximations (INLA), a methodology in which grid-based methods are used, is challenging. Motivated by this problem, we propose modifications to LDS-StM that improves the approximations and make it compatible with INLA, without sacrificing computational speed. We also present two examples to demonstrate that LDS-StM with modifications can outperform INLA's own grid approximation with respect to speed and accuracy. We also demonstrate the flexibility of the new approach for the approximation of multimodal marginals.Citation
Brown, P. T., Joshi, C., Joe, S., & Rue, H. (2021). A novel method of marginalisation using low discrepancy sequences for integrated nested Laplace approximations. Computational Statistics & Data Analysis, 157, 107147. doi:10.1016/j.csda.2020.107147Publisher
Elsevier BVarXiv
1911.09880Additional Links
https://arxiv.org/pdf/1911.09880ae974a485f413a2113503eed53cd6c53
10.1016/j.csda.2020.107147