Likelihood evaluations for stationary Gaussian time series are typically obtained via the Durbin-Levinson algorithm, which scales as O(n^2) in the number of time series observations. This package provides a "superfast" O(n log^2 n) algorithm written in C++, crossing over with Durbin-Levinson around n = 300. Efficient implementations of the score and Hessian functions are also provided, leading to superfast versions of inference algorithms such as Newton-Raphson and Hamiltonian Monte Carlo. The C++ code provides a Toeplitz matrix class packaged as a header-only library, to simplify low-level usage in other packages and outside of R.
| Version: | 1.0 |
| Depends: | R (≥ 3.0.0) |
| Imports: | stats, methods, Rcpp (≥ 0.12.7), fftw |
| LinkingTo: | Rcpp, RcppEigen |
| Suggests: | knitr, rmarkdown, testthat, mvtnorm, numDeriv |
| Published: | 2017-07-05 |
| Author: | Yun Ling [aut], Martin Lysy [aut, cre] |
| Maintainer: | Martin Lysy <mlysy at uwaterloo.ca> |
| License: | GPL-3 |
| NeedsCompilation: | yes |
| SystemRequirements: | FFTW (>= 3.1.2) |
| CRAN checks: | SuperGauss results |
| Reference manual: | SuperGauss.pdf |
| Vignettes: |
SuperGauss |
| Package source: | SuperGauss_1.0.tar.gz |
| Windows binaries: | r-devel: SuperGauss_1.0.zip, r-release: SuperGauss_1.0.zip, r-oldrel: SuperGauss_1.0.zip |
| OS X El Capitan binaries: | r-release: SuperGauss_1.0.tgz |
| OS X Mavericks binaries: | r-oldrel: SuperGauss_1.0.tgz |
Please use the canonical form https://CRAN.R-project.org/package=SuperGauss to link to this page.