Time-Decay Forecast Smoothing
Dynamic adjustment layer that re-weights prior EOQ history by recency, capturing momentum and velocity signals closer to quarter end.
Akash Sharma
Primary Owner
Production
Deployed
The Problem
Static average-based priors weighted all historical quarters equally, ignoring recent pipeline momentum or slowdown signals that are most predictive close to quarter end.
What Was Built
Built a time-decay adjustment layer that exponentially down-weights older EOQ observations, emphasizing recent quarters. Incorporates momentum, velocity, and capacity features. Deployed as a production adjustment layer on top of the base XGBoost forecast.
Business Impact
Improved late-quarter forecast accuracy by capturing recency signals, deployed to production as part of the Revenue Forecasting Platform.
Related Revenue Forecasting Projects
Revenue Forecasting Platform
Production ML system predicting pipeline and booking outcomes at daily frequency across current and future quarters.
Seasonal Index Forecast Adjustment
Statistical fallback layer using day-of-quarter index averages when ML model signals are insufficient.
Model Explainability with SHAP
Dual-algorithm explainability infrastructure: SHAP for tree models, coefficient contribution for linear models, with BigQuery logging.