Skip to contents

Fits a Bayesian hierarchical AR(1) state-space model relating global mean surface temperature (GMST) to a reduced paleoclimate proxy network, instrumental temperature observations, and radiative forcings (greenhouse gases, volcanic aerosols, and solar irradiance). See vignette("baygmst-intro", package = "BayGMST") for a worked end-to-end example, and vignette("baygmst-model", package = "BayGMST") for the full statistical model description, which builds on the reduced-proxy Bayesian framework of Barboza et al. (2014, 2019) and Wang (2020).

Model fitting requires CmdStan

fit_baygmst() and cv_baygmst() compile and sample a 'Stan' model via the instantiate and cmdstanr packages, which in turn require a working CmdStan installation on the user's machine. cmdstanr is not on CRAN; install it from https://stan-dev.r-universe.dev, then run cmdstanr::install_cmdstan(). BayGMST compiles its Stan models when it is installed, and only if cmdstanr and CmdStan are available at that moment, so install them first, or reinstall BayGMST after adding them. Use instantiate::stan_cmdstan_exists() to check whether CmdStan is available. reduce_proxies() and transform_forcings() do not require CmdStan.

Provenance

This package was restructured from a collection of top-level analysis scripts; see NEWS.md for exactly what changed (fixed bugs, dropped dead code, and modeling decisions reviewed by the authors) and system.file("legacy-scripts", package = "BayGMST") for the original, unmodified scripts.

Author

Maintainer: Nick McKay nick@nau.edu

Authors: