statsforecast_polars_ensemble_pipeline - Univariate Time Series Forecasting Pipeline
Execute univariate time series forecasting pipeline using StatsForecast and Polars with cross-validation, ensemble models, and formatted output.
Tags
Updated: 2026-05-11Capabilities
Typical Inputs
Typical Outputs
What this skill does
- concatenate identifiers
- rename columns
- filter data
- sort by date
- initialize models
- perform cross-validation
- apply constraints
- calculate ensemble
- calculate WMAPE
- calculate bias
- remove outliers
- generate forecasts
- round values
- split identifiers
- format output
Inputs
- DataFrame with MaterialID
- DataFrame with SalesOrg
- DataFrame with DistrChan
- DataFrame with CL4
- DataFrame with WeekDate
- DataFrame with OrderQuantity
Outputs
- forecast message
- metrics message
- forecast file
- metrics file
Requirements
- Polars
- StatsForecast
- Python