# robjhyndman/forecast

Forecasting Functions for Time Series and Linear Models

Repository: https://github.com/robjhyndman/forecast
Canonical: https://ross.abutalabs.com/products/forecast
Homepage: http://pkg.robjhyndman.com/forecast
Language: R
License Family: other
Topics: forecasting, forecast, cran, r
Last push: 2026-08-23T22:35:09+00:00

## Health v2 (maintenance only)
Score: 87/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 63, longevity 100
- inputs: {"age_days": 5246, "days_push": 10, "days_rel": 168, "gap_med": 34, "n_releases_24m": 4}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1174, forks 337 (observed 2026-08-28T04:03:52.252768+00:00)

## What it is
The R package forecast provides methods and tools for displaying and analysing univariate time series forecasts, including exponential smoothing via state space models and automatic ARIMA modelling. It is a mature, widely used CRAN package maintained by Rob Hyndman, with a tidyverse-oriented successor called fable.

## Use cases
- forecast future values of a time series in R
- automatically fit an ARIMA model to time series data
- exponential smoothing forecasts with ETS state space models
- decompose a seasonal time series with STL and forecast it
- forecast high-frequency or complex seasonal data with TBATS
- plot time series forecasts with ggplot2 autoplot
- estimate prediction intervals for time series forecasts

## When to choose
- you work in R and need classical statistical time series forecasting models like ARIMA, ETS, STL, or TBATS
- you want automatic model selection for univariate forecasting
- you need a battle-tested, well-documented package backed by an authoritative forecasting textbook
- you are following Hyndman's fpp2/fpp3 forecasting textbook examples

## When to avoid
- you prefer a tidyverse-style workflow - consider the fable package instead
- you need deep learning or neural network based forecasting
- you work primarily in Python rather than R
- you need multivariate or cross-sectional hierarchical forecasting beyond this package's scope

## Facets
- artifact type: library
- maturity: stable
- function: data-science, math, data-visualization
- domain: data-science, time-series, analytics
- platform: cross-platform
- tags: forecasting, time-series, arima, ets, exponential-smoothing, stl, tbats, cran, statistics, r

## Member repositories
- robjhyndman/forecast (main) score 87

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:52.252768+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:28:19.285597+00:00, confidence not recorded.
  - readme: https://github.com/robjhyndman/forecast (fetched 2026-08-28T04:03:52.252768+00:00, sha bfb6785bf239)
  - homepage: http://pkg.robjhyndman.com/forecast (fetched 2026-08-29T12:33:26.830284+00:00, sha 66bfe28afef1)
  - site_page: https://pkg.robjhyndman.com/forecast/news/index.html (fetched 2026-08-29T12:33:26.839781+00:00, sha c56a4f7c2c77)
  - site_page: https://pkg.robjhyndman.com/forecast/authors.html (fetched 2026-08-29T12:33:26.843059+00:00, sha f449ecd15a69)
- Data as of 2026-08-30T08:39:29.467469+00:00.