skforecast

Configures TimeSeriesFold parameters for backtesting based on deployment scenarios. Maps business requirements (retraining frequency, forecast horizon, data budget) to cross-validation strategy parameters. Use when the user describes how they plan to deploy or evaluate a model.

Complete constructor signatures 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.

Forecasts time series using recurrent neural networks (RNN, LSTM, GRU) with ForecasterRnn and the create_and_compile_model helper. Covers model architecture, training, and multi-series deep learning. Use when the user wants to use deep learning / neural networks for time series forecasting.

Creates features for time series forecasting: calendar features with skforecast's `CalendarFeatures` (cyclical, onehot, or spline encoding) — either delegated to the forecaster via the `calendar_features` parameter or built manually as exog — holiday distance features with `calculate_distance_from_holiday`, rolling statistics with `RollingFeatures`, differencing, and categorical exogenous variables. Use when the user wants to improve model accuracy through feature engineering or asks about exogenous variable creation.

Forecasts multiple time series simultaneously using a global model with ForecasterRecursiveMultiSeries or ForecasterDirectMultiVariate. Covers data formats, encoding, per-series transformers, and multi-series backtesting. Use when the user has two or more related time series.

Forecasts a single time series using ForecasterRecursive or ForecasterDirect. Covers data preparation, model creation, training, prediction, backtesting, and prediction intervals. Use when the user needs to predict future values of one time series.

Zero-shot time series forecasting with pre-trained foundation models (Amazon Chronos-2, Google TimesFM 2.5, Salesforce Moirai-2, Soda-INRIA TabICL, Prior Labs TabPFN-TS, The Forecasting Company T0) via ForecasterFoundation and FoundationModel. Covers single and multi-series workflows, exogenous variables, prediction intervals / quantiles, and backtesting. Use when the user wants forecasts without task-specific training, cold-start baselines, or pre-trained generalist models.

Optimizes forecaster hyperparameters using grid search, random search, or Bayesian search (Optuna). Covers single-series and multi-series search, cross-validation configuration, and search space definition. Use when the user wants to find the best model configuration.

Generates prediction intervals for time series forecasts using bootstrapping, conformal prediction, or built-in statistical model intervals. Covers interval configuration, residual management, and calibration. Use when the user needs uncertainty quantification for forecasts.

Forecasts time series using classical statistical models (ARIMA, SARIMAX, ETS, ARAR) wrapped in ForecasterStats. Covers model selection, Auto-ARIMA, backtesting statistical models, and parameter tuning. Use when the user wants traditional statistical forecasting methods.

Diagnoses and fixes common errors when using skforecast, especially mistakes frequently made by LLMs generating skforecast code. Covers deprecated imports, wrong function names, missing parameters, and data format issues. Use when generated code produces errors or unexpected results.

Guides selection of the appropriate skforecast forecaster based on the user's data characteristics and requirements. Provides a decision matrix mapping use cases to forecaster classes. Use when the user is unsure which forecaster to use or asks for a recommendation.

Guides selection of the most appropriate evaluation metric(s) for a forecasting task based on forecaster type, prediction output type, data characteristics, and multi-series aggregation needs. Use when the user asks which metric to use, how to evaluate forecast quality, or needs help configuring the `metric` parameter in backtesting or hyperparameter search.

Analyzes time series dynamics with the fast skforecast.stats functions acf, pacf and calculate_lag_autocorrelation. Covers reading ACF/PACF patterns to identify AR/MA orders and seasonality, ranking lags by partial autocorrelation, and feeding the result to the lags argument of any skforecast forecaster. Use when the user wants to understand the dynamics of a series, choose a candidate set of lags before hyperparameter tuning, or replace a slow statsmodels acf/pacf call.

Detects data drift in time series forecasting pipelines using RangeDriftDetector and PopulationDriftDetector. Covers range-based out-of-range detection and statistical distribution tests. Use when the user wants to monitor model reliability in production.

Selects the most relevant lags, window features, and exogenous variables using sklearn feature selectors (RFECV, SelectFromModel). Covers single-series and multi-series selection with force inclusion and subsampling. Use when the user has many features and wants to identify the most important ones.