Forecasting Inflation with Microdata: An Adaptive Machine Learning Approach

2026-07-14

AI Summary

This study addresses the challenge of improving macroeconomic inflation forecasts in non-stationary environments, particularly following major economic shocks, by leveraging heterogeneity in micro-level price dynamics. The authors propose an adaptive machine learning framework that encodes the distribution of price changes into high-dimensional features and employs gradient-boosted trees to generate micro-level predictions. A novel scanning statistical test is introduced to dynamically identify unknown-duration windows during which micro signals exhibit predictive power; these signals are incorporated into the ensemble forecast only when deemed effective. Empirical results demonstrate that, during the period of heightened inflation volatility in the UK post-2020, the proposed method significantly outperforms univariate benchmarks and consistently enhances forecast accuracy across all horizons, thereby confirming the unique predictive value of micro data in the presence of structural shocks.

Abstract

Does microeconomic heterogeneity help to forecast aggregate inflation in a non-stationary environment? We develop a scan test for whether one forecast outperforms another, over an interval with unknown starting point and duration. To exploit any occasional forecasting power that the scan test detects, we design an adaptive machine learning pipeline. We encode the distribution of price changes into a high-dimensional vector, which we combine with a gradient boosted trees algorithm. We then combine this micro forecast with other benchmark forecasts, using an adaptive algorithm that makes use of the micro forecast only when it performs well. We apply the pipeline to UK microdata, with four main results. First, the micro forecast outperforms a univariate benchmark, but only in the volatile period after 2020. Second, the scan test detects periods of micro outperformance, so the micro forecast enters the combined forecast. Third, the combined forecast performs comparably to the univariate benchmark before 2020 and better at every horizon after 2020. Fourth, the value of microdata for the combined forecast materializes after 2020. We conclude that microdata are valuable for forecasting aggregate inflation, but only after large shocks.

Problem

• inflation forecasting
• microdata
• non-stationary environment
• forecast combination
• economic shocks

Innovation

• adaptive machine learning
• scan test
• microdata
• inflation forecasting
• gradient boosted trees