Prophet

Prophet is a procedure for forecasting time series data based on an additive model where non-linear trends are fit with yearly, weekly, and daily seasonality. It works best with time series that have strong seasonal effects and several seasons of historical data.

Specify a linear, logistic or flat trend.
Saturating minimum and carrying capacity of the logistic growth model.
Use a Fourier decomposition to fit the yearly seasonality ('Auto' means only when input data is longer than 2 years).
Use a Fourier decomposition to fit the weekly seasonality ('Auto' means only when input data is longer than 2 weeks).
Use a Fourier decomposition to fit the daily seasonality ('Auto' means only when input data is longer than 2 days).
Proportion of history in which trend changepoints will be estimated (between 0 and 1).
Potential changepoints are selected uniformly from the first Changepoint range proportion of the history.
Using a fixed random seed allows for reproducible result.