tsissm: Linear Innovations State Space Unobserved Components Model

Unobserved components time series model using the linear innovations state space representation (single source of error) with choice of error distributions and option for dynamic variance. Methods for estimation using automatic differentiation, automatic model selection and ensembling, prediction, filtering, simulation and backtesting. Based on the model described in Hyndman et al (2012) <doi:10.1198/jasa.2011.tm09771>.

Version:	1.0.1
Depends:	R (≥ 4.1.0), Rcpp (≥ 0.12.9), tsmethods (≥ 1.0.0)
Imports:	TMB (≥ 1.7.20), methods, tsaux, tsdistributions, zoo, xts, copula, flextable, data.table, sandwich, nloptr, RTMB, viridisLite, future, future.apply, progressr
LinkingTo:	Rcpp (≥ 0.12.9), TMB, RcppEigen
Suggests:	rmarkdown, tstests, knitr, testthat (≥ 3.0.0)
Published:	2025-04-28
DOI:	10.32614/CRAN.package.tsissm
Author:	Alexios Galanos [aut, cre]
Maintainer:	Alexios Galanos <alexios at 4dscape.com>
License:	GPL-2
URL:	https://github.com/tsmodels/tsissm, https://www.nopredict.com/packages/tsissm
NeedsCompilation:	yes
Materials:	NEWS
CRAN checks:	tsissm results

Documentation:

Reference manual:	tsissm.pdf
Vignettes:	Automatic Selection and Ensembling (source, R code) Filtering and Prediction (source, R code) Regular Spacings and Seasonality (source, R code) Variance Dynamics (source, R code) The Innovations State Space Model (source, R code)

Downloads:

Package source:	tsissm_1.0.1.tar.gz
Windows binaries:	r-devel: not available, r-release: not available, r-oldrel: not available
macOS binaries:	r-release (arm64): tsissm_1.0.1.tgz, r-oldrel (arm64): not available, r-release (x86_64): tsissm_1.0.1.tgz, r-oldrel (x86_64): tsissm_1.0.1.tgz

Linking:

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