| name | complete-api-reference |
| description | Provides complete constructor and method signatures for all skforecast forecasters, backtesting functions, search functions, cross-validation classes, preprocessing, feature selection, and drift detection. Use when the user needs exact parameter names, types, or defaults for any skforecast class or function.
|
Complete API Reference
When to Use
Use this when you need exact parameter names, types, defaults, or method signatures for any skforecast class or function.
Related skills
- Prerequisite:
choosing-a-forecaster (pick the class first; this skill only supplies its signature)
- Alongside: any workflow skill (they show the idiomatic usage, this one the exact arguments)
- Next:
troubleshooting-common-errors (when a signature looks right but the call still fails)
Quick Index
Forecaster Constructors
ForecasterRecursive — single series, recursive strategyForecasterRecursiveMultiSeries — multiple series, global modelForecasterDirect — single series, one model per stepForecasterDirectMultiVariate — multiple input series, one targetForecasterRecursiveClassifier — classification-basedForecasterStats — statistical models (ARIMA, ETS, SARIMAX, ARAR)ForecasterEquivalentDate — baseline using past offsetsForecasterRnn — deep learning (RNN/LSTM/GRU)ForecasterFoundation — zero-shot with foundation models (Chronos-2, TimesFM 2.5, Moirai-2, TabICL, TabPFN-TS, TFC-T0, Nori, TS-IC)FoundationModel — low-level foundation model wrapper used by ForecasterFoundation
Forecaster Methods
fit() — train the modelpredict() — generate point forecastspredict_interval() — generate prediction intervals
Model Selection
backtesting_forecaster — backtest single-series forecastersbacktesting_forecaster_multiseries — backtest multi-series forecastersbacktesting_stats — backtest statistical modelsbacktesting_foundation — backtest ForecasterFoundation (zero-shot)grid_search_forecaster / grid_search_forecaster_multiseries / grid_search_statsrandom_search_forecaster / random_search_forecaster_multiseries / random_search_statsbayesian_search_forecaster / bayesian_search_forecaster_multiseriesbayesian_search_foundation — tune ForecasterFoundation inference-time parametersgrid_search_equivalent_date — tune ForecasterEquivalentDate baselinesTimeSeriesFold — multi-step cross-validationOneStepAheadFold — fast one-step cross-validation
Feature Selection
select_features — single seriesselect_features_multiseries — multi-series
Drift Detection
RangeDriftDetector — lightweight range checkPopulationDriftDetector — statistical tests
Preprocessing
RollingFeatures — rolling window statisticsTimeSeriesDifferentiator — differencingCalendarFeatures — calendar features
References
Full constructor and method signatures for all public skforecast classes and
functions, split by domain:
- Forecaster constructors: See references/forecaster-constructors.md
- Forecaster methods (
fit(), predict(), predict_interval(), predict_quantiles(), predict_dist(), set_params(), set_lags(), set_out_sample_residuals(), availability matrix): See references/forecaster-methods.md
- Backtesting, search, cross-validation, feature selection: See references/model-selection-signatures.md
- Drift detection and preprocessing: See references/preprocessing-signatures.md
