Rolling Revenue Forecasting - Deployment

This workflow generate rolling revenue forecasts across regions and product lines, using a dynamic walk-forward approach that reflects real-world monthly forecasting. It imports a trained regression model and uses lag-based features to capture temporal revenue dependencies. In the rolling forecast scenario, predictions are recursively fed back into the dataset and the model to generate sequential forecasts for future periods. The workflow relies on historical monthly data of 2023, with the goal of generating monthly forecasts for 2024 across regions and products. Results and predictions are presented in an interactive dashboard.

Visualization

Generate Monthly Rolling Forecasts

• Implements walk-forward logic by generating lag features, applying the trained model, and recursively updating the dataset with new predictions.

Data Reading