Prophet
Prophet forecasts time series data based on an additive model with non-linear trends. It supports yearly, weekly, and daily seasonality, and accounts for holiday effects.
We implemented prophet as the agile-robust option in our Anomaly detection practice. We recommend that you use it with time series that exhibit strong seasonal effects, and contain several cycles of historical data.
For an in-depth discussion of the Prophet algorithm, see these external resources:
-
Prophet project ion Facebook
-
Forecasting at scale in PeerJ
