Sktime Forecasting

Introduction

The Sktime Forecasting integration provides specialized optimization capabilities for time series forecasting models. This integration handles the unique challenges of time series data, including proper temporal cross-validation and forecasting-specific parameter spaces.

Key Features

• Time series cross-validation: Proper temporal validation strategies
• Forecasting model support: Integration with sktime forecasters
• Horizon optimization: Multi-step ahead forecasting optimization
• Seasonal parameter tuning: Seasonal model parameter optimization

Basic Usage

Time Series Cross-Validation

Seasonal Model Optimization

Multi-Step Forecasting

References

• Sktime documentation: https://www.sktime.org/
• Time series cross-validation: https://www.sktime.org/en/latest/api_reference/auto_generated/sktime.forecasting.model_selection.temporal_train_test_split.html