smoots: Nonparametric Estimation of the Trend and Its Derivatives in TS

The nonparametric trend and its derivatives in equidistant time series (TS) with short-memory stationary errors can be estimated. The estimation is conducted via local polynomial regression using an automatically selected bandwidth obtained by a built-in iterative plug-in algorithm or a bandwidth fixed by the user. A Nadaraya-Watson kernel smoother is also built-in as a comparison. The methods of the package are described in Feng, Y., and Gries, T., (2017) <>. A current version of the paper that is also referred to in the documentation of the functions is prepared for publication.

Version: 1.0.1
Depends: R (≥ 2.10)
Imports: stats, graphics
Suggests: knitr, rmarkdown, fGarch
Published: 2019-12-02
Author: Yuanhua Feng [aut] (Paderborn University, Germany), Dominik Schulz [aut, cre] (Paderborn University, Germany), Thomas Gries [ctb] (Paderborn University, Germany), Marlon Fritz [ctb] (Paderborn University, Germany), Sebastian Letmathe [ctb] (Paderborn University, Germany)
Maintainer: Dominik Schulz <schulzd at>
License: GPL-3
NeedsCompilation: no
Materials: README NEWS
In views: TimeSeries
CRAN checks: smoots results


Reference manual: smoots.pdf
Package source: smoots_1.0.1.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
macOS binaries: r-release: smoots_1.0.1.tgz, r-oldrel: smoots_1.0.1.tgz
Old sources: smoots archive


Please use the canonical form to link to this page.