Simulation-Augmented Multi-Step Split Conformal Prediction for Aggregated Forecasts
2026-06-15 • Machine Learning
Machine Learning
AI summaryⓘ
The authors focus on predicting overall totals and growth rates over time, like yearly sales amounts. They created a new method called SA-MSCP that uses simulations combined with past prediction mistakes to better estimate how uncertain these forecasts are. Their tests show this method gives more accurate confidence intervals compared to traditional approaches. This shows their approach is useful for understanding uncertainty in time-based aggregated predictions.
uncertainty quantificationaggregated forecastingtime seriesprediction intervalsconformal methodsbootstrapcross-validationsimulationgrowth rates
Authors
Andro Sabashvili
Abstract
We study uncertainty quantification for aggregated forecasting tasks such as annual totals and year-over-year growth rates. We propose SA-MSCP, a simulation-augmented multi-step split conformal method that generates future paths from cross-validated residuals using a block bootstrap and constructs prediction intervals from empirical quantiles. Experiments show that SA-MSCP improves empirical coverage over a simulated-path baseline for aggregated and growth-rate targets. Our results demonstrate that simulation-enhanced conformal calibration is an effective and general framework for uncertainty quantification in aggregated time-series forecasting.