English

Laplace and Saddlepoint Approximations in High Dimensions

Statistics Theory 2023-11-09 v3 Methodology Statistics Theory

Abstract

We examine the behaviour of the Laplace and saddlepoint approximations in the high-dimensional setting, where the dimension of the model is allowed to increase with the number of observations. Approximations to the joint density, the marginal posterior density and the conditional density are considered. Our results show that under the mildest assumptions on the model, the error of the joint density approximation is O(p4/n)O(p^4/n) if p=o(n1/4)p = o(n^{1/4}) for the Laplace approximation and saddlepoint approximation, and O(p3/n)O(p^3/n) if p=o(n1/3)p = o(n^{1/3}) under additional assumptions on the second derivative of the log-likelihood. Stronger results are obtained for the approximation to the marginal posterior density.

Cite

@article{arxiv.2107.10885,
  title  = {Laplace and Saddlepoint Approximations in High Dimensions},
  author = {Yanbo Tang and Nancy Reid},
  journal= {arXiv preprint arXiv:2107.10885},
  year   = {2023}
}

Comments

56 pages

R2 v1 2026-06-24T04:26:37.547Z