Implements a procedure for forecasting time series data based on an additive model where non-linear trends are fit with yearly, weekly, and daily seasonality, plus holiday effects. It works best with time series that have strong seasonal effects and several seasons of historical data. Prophet is robust to missing data and shifts in the trend, and typically handles outliers well.
| Version: | 0.3.0.1 |
| Depends: | R (≥ 3.2.3), Rcpp (≥ 0.12.0) |
| Imports: | dplyr (≥ 0.5.0), dygraphs (≥ 1.1.1.4), extraDistr, ggplot2, grid, rstan (≥ 2.14.0), scales, stats, tidyr (≥ 0.6.1), xts |
| Suggests: | knitr, testthat, readr |
| Published: | 2018-06-15 |
| Author: | Sean Taylor [cre, aut], Ben Letham [aut] |
| Maintainer: | Sean Taylor <sjt at fb.com> |
| License: | BSD_3_clause + file LICENSE |
| NeedsCompilation: | yes |
| SystemRequirements: | C++11 |
| In views: | TimeSeries |
| CRAN checks: | prophet results |
| Reference manual: | prophet.pdf |
| Vignettes: |
Quick Start Guide to Using Prophet |
| Package source: | prophet_0.3.0.1.tar.gz |
| Windows binaries: | r-devel: prophet_0.3.0.1.zip, r-release: prophet_0.3.0.1.zip, r-oldrel: prophet_0.3.0.1.zip |
| OS X binaries: | r-release: prophet_0.3.0.1.tgz, r-oldrel: prophet_0.3.0.1.tgz |
| Old sources: | prophet archive |
Please use the canonical form https://CRAN.R-project.org/package=prophet to link to this page.