Age-specific demographic modeling and forecasting: Rolling window, expanding window, or both?

2026-07-11

AI Summary

This study addresses the impact of windowing strategies on the accuracy of age-specific population forecasts, particularly in modeling mortality and fertility rates. Through empirical comparisons across multiple countries, it evaluates rolling and expanding window approaches and proposes a simple ensemble method that assigns equal weights to both without requiring prior knowledge of their relative performance. The findings indicate that the superiority of a single windowing strategy tends to remain consistent across different forecast horizons. Moreover, the proposed equal-weight combination demonstrates robust and competitive predictive performance overall, offering a practical and effective alternative for real-world forecasting scenarios where optimal window selection is uncertain or unknown.

Abstract

Rolling and expanding windows are widely used in age-specific demographic modeling and forecasting. Building on these approaches, we propose a simple combination method that assigns equal weight to the forecasts from both schemes. Our focus is on evaluating and comparing the forecast accuracy of the two window types in modeling age-specific mortality and fertility. Based on the multi-country comparison, the superior performance of one method often persists across different forecast horizons. In the absence of prior information, our combined approach offers a robust and practical alternative.

Problem

• age-specific demographic modeling
• forecasting
• rolling window
• expanding window
• forecast accuracy

Innovation

• rolling window
• expanding window
• combined forecasting
• age-specific demographic modeling
• forecast accuracy