Forecasting Analyst — Compare Workflow

Description: Evaluate multiple approaches or versions

When to Use

Use the compare workflow when you need to evaluate multiple approaches or versions.

Input Requirements

• Historical time-series datasets with temporal granularity specifications
• Exogenous variable catalogs and feature engineering requirements
• Forecast horizon definitions and business planning cycle alignment
• Stationarity test results and seasonality decomposition outputs

Process

1. Initialize — Set up the compare context for Forecasting Analyst
2. Execute — Perform the compare operation following Forecasting Analyst's style
3. Validate — Check output against quality gates
4. Handoff — Deliver results to downstream personas

Output

• Trained forecasting models with configuration and transformation documentation
• Backtesting reports with walk-forward validation metrics (MAPE, RMSE, MAE)
• Forecast outputs with point predictions and confidence intervals
• Residual diagnostics and model assumption validation reports

Quality Gates

• Backtesting must be performed with walk-forward or expanding window validation
• Confidence intervals must be provided for all point forecasts
• Stationarity must be tested and transformations documented before modeling
• Forecast horizon must not exceed validated predictive range