An implementation of a number of Global Trend models for time series forecasting that are Bayesian generalizations and extensions of some Exponential Smoothing models. The main differences/additions include 1) nonlinear global trend, 2) Student-t error distribution, and 3) a function for the error size, so heteroscedasticity. The methods are particularly useful for short time series. When tested on the well-known M3 dataset, they are able to outperform all classical time series algorithms. The models are fitted with MCMC using the 'rstan' package.
| Version: | 0.2-3 |
| Depends: | R (≥ 3.4.0),Rcpp (≥ 0.12.0), methods,rstantools,forecast,truncnorm |
| Imports: | rstan (≥ 2.26.0),sn |
| LinkingTo: | StanHeaders (≥ 2.26.0),rstan (≥ 2.26.0),BH (≥ 1.66.0),Rcpp (≥ 0.12.0),RcppEigen (≥ 0.3.3.3.0),RcppParallel (≥5.0.2) |
| Suggests: | doParallel,foreach,knitr,rmarkdown,Mcomp,RODBC,dplyr,ggplot2 |
| Published: | 2025-04-30 |
| DOI: | 10.32614/CRAN.package.Rlgt |
| Author: | Slawek Smyl [aut], Christoph Bergmeir [aut, cre], Erwin Wibowo [aut], To Wang Ng [aut], Xueying Long [aut], Alexander Dokumentov [aut], Daniel Schmidt [aut], Trustees of Columbia University [cph] (tools/make_cpp.R, R/stanmodels.R) |
| Maintainer: | Christoph Bergmeir <christoph.bergmeir at monash.edu> |
| License: | GPL-3 |
| URL: | https://github.com/cbergmeir/Rlgt |
| NeedsCompilation: | yes |
| SystemRequirements: | GNU make |
| Materials: | ChangeLog |
| In views: | TimeSeries |
| CRAN checks: | Rlgt results |
| Reference manual: | Rlgt.html ,Rlgt.pdf |
| Vignettes: | Global Trend Models - LGT, SGT, and S2GT (source,R code) Getting Started with Global Trend Models (source,R code) |
| Package source: | Rlgt_0.2-3.tar.gz |
| Windows binaries: | r-devel:Rlgt_0.2-3.zip, r-release:Rlgt_0.2-3.zip, r-oldrel:Rlgt_0.2-3.zip |
| macOS binaries: | r-release (arm64):Rlgt_0.2-3.tgz, r-oldrel (arm64):Rlgt_0.2-3.tgz, r-release (x86_64):Rlgt_0.2-3.tgz, r-oldrel (x86_64):Rlgt_0.2-3.tgz |
| Old sources: | Rlgt archive |
Please use the canonical formhttps://CRAN.R-project.org/package=Rlgtto link to this page.