The R Journal: article published in 2020, volume 12:1

Skew-t Expected Information Matrix Evaluation and Use for Standard Error Calculations
R. Douglas Martin, Chindhanai Uthaisaad and Daniel Z. Xia , The R Journal (2020) 12:1, pages 188-205.

Abstract Skew-t distributions derived from skew-normal distributions, as developed by Azzalini and several co-workers, are popular because of their theoretical foundation and the availability of computational methods in the R package sn. One difficulty with this skew-t family is that the elements of the expected information matrix do not have closed form analytic formulas. Thus, we developed a numerical integration method of computing the expected information matrix in the R package skewtInfo. The accuracy of our expected information matrix calculation method was confirmed by comparing the result with that obtained using an observed information matrix for a very large sample size. A Monte Carlo study to evaluate the accuracy of the standard errors obtained with our expected information matrix calculation method, for the case of three realistic skew-t parameter vectors, indicates that use of the expected information matrix results in standard errors as accurate as, and sometimes a little more accurate than, use of an observed information matrix.

Received: 2019-06-07; online 2020-09-10, supplementary material, (2.5 KiB)
CRAN packages: sn
CRAN Task Views implied by cited CRAN packages: Distributions, Multivariate

This article and supplementary materials are licensed under a Creative Commons Attribution 4.0 International license.

@article{RJ-2020-019,
  author = {R. Douglas Martin and Chindhanai Uthaisaad and Daniel Z. Xia},
  title = {{Skew-t Expected Information Matrix Evaluation and Use for
          Standard Error Calculations}},
  year = {2020},
  journal = {{The R Journal}},
  doi = {10.32614/RJ-2020-019},
  url = {https://doi.org/10.32614/RJ-2020-019},
  pages = {188--205},
  volume = {12},
  number = {1}
}