walker: Bayesian Regression with Time-Varying Coefficients

Bayesian dynamic regression models where the regression coefficients can vary over time as random walks. Gaussian, Poisson, and binomial observations are supported. The Markov chain Monte Carlo computations are done using Hamiltonian Monte Carlo provided by Stan, using a state space representation of the model in order to marginalise over the coefficients for efficient sampling. For non-Gaussian models, walker uses the importance sampling type estimators based on approximate marginal MCMC as in Vihola, Helske, Franks (2017, <arXiv:1609.02541>).

Version: 0.3.1-1
Depends: R (≥ 3.4.0), Rcpp (≥ 0.12.9), bayesplot, rstan (≥ 2.18.1)
Imports: dplyr, ggplot2, KFAS, methods
LinkingTo: StanHeaders (≥ 2.18.0), rstan (≥ 2.18.1), BH (≥ 1.66.0), Rcpp (≥ 0.12.9), RcppArmadillo, RcppEigen (≥
Suggests: diagis, gridExtra, knitr (≥ 1.11), rmarkdown (≥ 0.8.1), testthat
Published: 2020-01-23
Author: Jouni Helske ORCID iD [aut, cre]
Maintainer: Jouni Helske <jouni.helske at iki.fi>
BugReports: https://github.com/helske/walker/issues
License: GPL (≥ 3)
URL: https://github.com/helske/walker
NeedsCompilation: yes
SystemRequirements: C++14, GNU make
Citation: walker citation info
Materials: README
CRAN checks: walker results


Reference manual: walker.pdf
Vignettes: Efficient Bayesian generalized linear models with time-varying coefficients
Package source: walker_0.3.1-1.tar.gz
Windows binaries: r-prerelease: walker_0.3.1-1.zip, r-release: walker_0.3.1-1.zip, r-oldrel: walker_0.3.1-1.zip
macOS binaries: r-prerelease: walker_0.3.1-1.tgz, r-release: walker_0.2.5.tgz, r-oldrel: walker_0.2.4-1.tgz
Old sources: walker archive


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