Implements Bayesian dynamic factor analysis with 'Stan'. Dynamic factor analysis is a dimension reduction tool for multivariate time series. 'bayesdfa' extends conventional dynamic factor models in several ways. First, extreme events may be estimated in the latent trend by modeling process error with a student-t distribution. Second, autoregressive and moving average components can be optionally included. Third, the estimated dynamic factors can be analyzed with hidden Markov models to evaluate support for latent regimes.
| Version: | 0.1.0 |
| Depends: | R (≥ 3.4.0), Rcpp (≥ 0.12.0) |
| Imports: | rstan (≥ 2.16.0), ggplot2 (≥ 2.2.0), loo, dplyr (≥ 0.7.0), methods, rlang, reshape2 |
| LinkingTo: | StanHeaders (≥ 2.16.0), rstan (≥ 2.16.0), BH (≥ 1.62.0), Rcpp (≥ 0.12.0), RcppEigen |
| Suggests: | testthat, parallel, knitr, rmarkdown, MARSS |
| Published: | 2018-09-19 |
| Author: | Eric J. Ward [aut, cre], Sean C. Anderson [aut], Luis A. Damiano [aut], Mary E. Hunsicker, [ctb], Mike A. Litzow [ctb], Trustees of Columbia University [cph] |
| Maintainer: | Eric J. Ward <eric.ward at noaa.gov> |
| BugReports: | https://github.com/fate-ewi/bayesdfa/issues |
| License: | GPL (≥ 3) |
| URL: | https://github.com/fate-ewi/bayesdfa |
| NeedsCompilation: | yes |
| Materials: | NEWS |
| CRAN checks: | bayesdfa results |
| Reference manual: | bayesdfa.pdf |
| Vignettes: |
Estimating latent trends with bayesdfa |
| Package source: | bayesdfa_0.1.0.tar.gz |
| Windows binaries: | r-devel: bayesdfa_0.1.0.zip, r-release: bayesdfa_0.1.0.zip, r-oldrel: not available |
| OS X binaries: | r-release: not available, r-oldrel: not available |
Please use the canonical form https://CRAN.R-project.org/package=bayesdfa to link to this page.