Package: rjd3filters Type: Package Title: Trend-Cycle Extraction with Linear Filters with 'JDemetra+ 3.x' Version: 2.4.0 Authors@R: c( person("Alain", "Quartier-la-Tente", role = c("aut", "cre", "cph"), email = "alain.quartier@yahoo.fr", comment = c(ORCID = "0000-0001-7890-3857")), person("Jean", "Palate", role = c("aut"), email = "palatejean@gmail.com"), person("Tanguy", "Barthelemy", role = c("ctb"), email ="tanguy.barthelemy@insee.fr"), person("Anna", "Smyk", role = c("ctb"), email ="anna.smyk@insee.fr") ) Description: Build and apply symmetric and asymmetric moving averages (= linear filters) for trend-cycle extraction. In particular, it implements several modern approaches for real-time estimates from the viewpoint of revisions and time delay in detecting turning points. It includes the local polynomial approach of Proietti and Luati (2008), the Reproducing Kernel Hilbert Space (RKHS) of Dagum and Bianconcini (2008) and the Fidelity-Smoothness-Timeliness approach of Grun-Rehomme, Guggemos, and Ladiray (2018). It is based on 'Java' libraries developed in 'JDemetra+' (), time series analysis software. Depends: R (>= 4.1.0) Imports: rJava (>= 1.0-6), methods, MASS, graphics, stats, utils, rjd3jars (>= 0.0.5), rjd3toolkit (>= 3.7.1) SystemRequirements: Java (>= 21) License: EUPL LazyData: TRUE URL: https://github.com/rjdverse/rjd3filters, https://rjdverse.github.io/rjd3filters/ BugReports: https://github.com/rjdverse/rjd3filters/issues Suggests: knitr, rmarkdown, spelling Encoding: UTF-8 Config/roxygen2/version: 8.1.0 Language: en-GB NeedsCompilation: no Packaged: 2026-10-10 14:10:41 UTC; root Author: Alain Quartier-la-Tente [aut, cre, cph] (ORCID: ), Jean Palate [aut], Tanguy Barthelemy [ctb], Anna Smyk [ctb] Maintainer: Alain Quartier-la-Tente Config/pak/sysreqs: default-jdk Repository: https://cran.r-universe.dev Date/Publication: 2026-10-10 13:53:14 UTC RemoteUrl: https://github.com/cran/rjd3filters RemoteRef: HEAD RemoteSha: d1f0dae69968721e8aae8d5c4e30794994a083c5