From Many Models, One: Macroeconomic Forecasting with Reservoir Ensembles

2025-12-15

AI Summary

Single-model approaches suffer from insufficient generalization capability in macroeconomic time series forecasting. Method: This paper proposes a dynamic ensemble method based on multi-frequency echo state networks (MFESNs), incorporating Hedge and Follow-the-Leader online learning algorithms to construct a data-dependent, real-time weighting mechanism for fusing multiple MFESN reservoir models. Contribution/Results: We extend online learning theoretical guarantees—specifically, regret bounds—to data-dependent settings for the first time, establishing the first MFESN ensemble framework with rigorous regret analysis. Empirical evaluations across multiple macroeconomic forecasting tasks demonstrate that the proposed method significantly outperforms standalone MFESN models, achieving higher prediction accuracy and enhanced robustness.

Technology Category

• Machine Learning: Ensemble Methods
• Search and Optimization: Learning to Search
• Multiagent Systems: Multiagent Learning

Application Category

• Economics, Online Markets and Human Computation: Social networks and social learning
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Model combination is a powerful approach to achieve superior performance with a set of models than by just selecting any single one. We study both theoretically and empirically the effectiveness of ensembles of Multi-Frequency Echo State Networks (MFESNs), which have been shown to achieve state-of-the-art macroeconomic time series forecasting results (Ballarin et al., 2024a). Hedge and Follow-the-Leader schemes are discussed, and their online learning guarantees are extended to the case of dependent data. In applications, our proposed Ensemble Echo State Networks show significantly improved predictive performance compared to individual MFESN models.

Problem

• Improves macroeconomic forecasting using ensemble models
• Extends online learning guarantees to dependent data
• Enhances predictive performance over individual neural networks

Innovation

• Ensembles of Multi-Frequency Echo State Networks for forecasting
• Hedge and Follow-the-Leader online learning schemes
• Extended guarantees for dependent data in model combination