MSIMST: Bayesian Monotonic Single-Index Regression Model with the Skew-T Likelihood

Incorporates a Bayesian monotonic single-index mixed-effect model with a multivariate skew-t likelihood, specifically designed to handle survey weights adjustments. Features include a simulation program and an associated Gibbs sampler for model estimation. The single-index function is constrained to be monotonic increasing, utilizing a customized Gaussian process prior for precise estimation. The model assumes random effects follow a canonical skew-t distribution, while residuals are represented by a multivariate Student-t distribution. Offers robust Bayesian adjustments to integrate survey weight information effectively.

Version: 1.1
Depends: R (≥ 3.4.0)
Imports: MASS (≥ 7.3-58.4), Rcpp (≥ 1.0.12), mvtnorm (≥ 1.2-4), fields (≥ 15.2), parallel (≥ 4.3.0), truncnorm (≥ 1.0-9), Rdpack (≥ 2.6)
LinkingTo: Rcpp, RcppArmadillo
Suggests: lattice (≥ 0.21-8), HDInterval (≥ 0.2.4), latex2exp (≥ 0.9.6), posterior (≥ 1.5.0)
Published: 2024-09-16
DOI: 10.32614/CRAN.package.MSIMST
Author: Qingyang Liu ORCID iD [aut, cre], Debdeep Pati [aut], Dipankar Bandyopadhyay [aut]
Maintainer: Qingyang Liu <rh8liuqy at gmail.com>
License: GPL (≥ 3)
URL: https://github.com/rh8liuqy/MSIMST
NeedsCompilation: yes
Materials: README NEWS
CRAN checks: MSIMST results

Documentation:

Reference manual: MSIMST.pdf
Vignettes: Robust Statistical Modeling for Quantifying Periodontal Disease: A Single Index Mixed-Effects Approach with Skewed Random Effects and Heavy-Tailed Residuals (source)

Downloads:

Package source: MSIMST_1.1.tar.gz
Windows binaries: r-devel: MSIMST_1.1.zip, r-release: not available, r-oldrel: MSIMST_1.1.zip
macOS binaries: r-release (arm64): MSIMST_1.1.tgz, r-oldrel (arm64): MSIMST_1.1.tgz, r-release (x86_64): not available, r-oldrel (x86_64): not available

Linking:

Please use the canonical form https://CRAN.R-project.org/package=MSIMST to link to this page.