tsforecast_forecast | Fits statistical models to one series (y) or every series of a panel (data: id, time, value) and forecasts h periods ahead: any mix of automatic ARIMA, ETS, CES, Theta, TBATS, MSTL, boosted decomposition, structural, intermittent-demand, GARCH, benchmark and regression models (tsforecast_models lists them and their parameters), with native or conformal prediction intervals, numeric exogenous regressors (future_exog) and a fallback model for series a model cannot fit (its forecasts then fill that model’s column). | |
tsforecast_models | Lists the forecasting models any multi-model tool accepts: each model key, the family tool that also returns its fitted summary, its default column name, a one-line description and what it supports (exogenous regressors, native intervals, transfer to new data, simulation), with every parameter’s type, default, allowed range and meaning. | |
tsforecast_forward | Fits the models on one history (data or y) and applies them, parameters unchanged, to new data (new_data or new_y): the same series with more observations (update a forecast without re-estimating) or other series (forecast a short or new series with a model learned elsewhere). | |
