Prophet

Prophet, created and maintained by Facebook's (Meta's) Core Data Science team as open source, is a time series forecasting tool using an additive model that fits non-linear trends with yearly, weekly, and daily seasonality plus holiday effects. It's specifically designed to be robust to missing data, trend shifts, and outliers, and is available in both R and Python, using Stan under the hood to deliver forecasts in seconds with tunable parameters for domain-specific adjustment. It has been used extensively inside Meta itself for production forecasting, a real, checkable pedigree beyond typical open-source projects. Completely free and open source, available on CRAN (R) and PyPI (Python). Good fit for a data team forecasting a metric with strong seasonal patterns (sales, traffic, demand) that wants fast, automated forecasting without deep time-series modeling expertise, while still supporting manual tuning when needed.