Predicting the Future by Retrieving the Past

2025/11/08 by Du, Dazhao, Han, Tao, Guo, Song
Computer Science · Decision Sciences · #Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #Forecasting Techniques and Applications #Machine Learning (cs.LG) #Stock Market Forecasting Methods #Time Series Analysis and Forecasting

paper · doi:10.48550/arxiv.2511.05859

openalex publication_date 2025/11/08 · openalex created_date 2025/11/12 · openalex updated_date 2026/07/28

Abstract

Deep learning models such as MLP, Transformer, and TCN have achieved remarkable success in univariate time series forecasting, typically relying on sliding window samples from historical data for training. However, while these models implicitly compress historical information into their parameters during training, they are unable to explicitly and dynamically access this global knowledge during inference, relying only on the local context within the lookback window. This results in an underutilization of rich patterns from the global history. To bridge this gap, we propose Predicting the Future by Retrieving the Past (PFRP), a novel approach that explicitly integrates global historical data to enhance forecasting accuracy. Specifically, we construct a Global Memory Bank (GMB) to effectively store and manage global historical patterns. A retrieval mechanism is then employed to extract similar patterns from the GMB, enabling the generation of global predictions. By adaptively combining these global predictions with the outputs of any local prediction model, PFRP produces more accurate and interpretable forecasts. Extensive experiments conducted on seven real-world datasets demonstrate that PFRP significantly enhances the average performance of advanced univariate forecasting models by 8.4%. Codes can be found in https://github.com/ddz16/PFRP.

Citations

• Sundial: A Family of Highly Capable Time Series Foundation Models
• TimeCMA: Towards LLM-Empowered Multivariate Time Series Forecasting via Cross-Modality Alignment
• SparseTSF: Modeling Long-term Time Series Forecasting with 1k Parameters
• TimeXer: Empowering Transformers for Time Series Forecasting with Exogenous Variables
• MOMENT: A Family of Open Time-series Foundation Models
• Unified Training of Universal Time Series Forecasting Transformers
• FengWu: Pushing the Skillful Global Medium-range Weather Forecast beyond 10 Days Lead
• One Fits All:Power General Time Series Analysis by Pretrained LM
• FEDformer: Frequency Enhanced Decomposed Transformer for Long-term Series Forecasting
• Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting
• Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting
• Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks
• A Simple Framework for Contrastive Learning of Visual Representations
• REALM: Retrieval-Augmented Language Model Pre-Training
• Temporal Fusion Transformers for Interpretable Multi-horizon Time Series\n Forecasting
• Temporal Fusion Transformers for interpretable multi-horizon time series forecasting
• N-BEATS: Neural basis expansion analysis for interpretable time series forecasting
• Attention Is All You Need
• DeepAR: Probabilistic Forecasting with Autoregressive Recurrent Networks
• Adam: A Method for Stochastic Optimization

Related

• Retrieval Augmented Time Series Forecasting
• DAM: Towards A Foundation Model for Time Series Forecasting
• What Matters in Deep Learning for Time Series Forecasting?
• Think Globally, Act Locally: A Deep Neural Network Approach to High-Dimensional Time Series Forecasting
• TreeDRNet:A Robust Deep Model for Long Term Time Series Forecasting