Forecast Collapse in Time-Series Foundation Models

2026-08-14 · Shu Wan, Miles Ma, Hank Zhu, Guangqi Liu, Stephen Wang, Qingsong Wen, Huan Liu arxiv

When forecasting hourly returns for 1,000 US equities, we observe an unexpected phenomenon: predictions become nearly flat and show poor stock ranking, as measured by cross-sectional correlation. We call this forecast collapse. Surprisingly, the phenomenon largely disappears when forecasting trading volume under the same setting. We investigate forecast collapse across time-series foundation models (TSFMs), twelve deep-learning forecasting models, and 97 public benchmark configurations, and find that it is closely tied to target predictability. We identify two distinct reasons behind it: low predictability limits the amplitude of calibrated point forecasts, while per-series objectives leave cross-series structure unidentified. These findings reveal a calibration-ranking tradeoff: optimizing squared error leads to flat predictions, whereas directly optimizing cross-sectional correlation improves ranking but can inflate forecast amplitude by more than an order of magnitude. To address this tradeoff, we introduce CalibRank, a simple objective that balances calibration and ranking. On Finance1K, CalibRank nearly triples cross-sectional correlation while keeping amplitude close to the target, and improves correlation on all tested models. Our results reveal a blind spot in conventional time-series evaluation: per-series metrics can hide failures in cross-series structure needed by downstream decisions.

📄 PDF Abstract BibTeX arXiv:2608.14106

Code (0)

등록된 구현이 없습니다.

Sundial: A Family of Highly Capable Time Series Foundation Models

2025-02-02 · Yong liu, Guo Qin, Zhiyuan Shi, Zhi Chen 외

We introduce Sundial, a family of native, flexible, and scalable time series foundation models. To predict the next-patch's distribution, we propose a TimeFlow Loss based on flow-matching, which facilitates native pre-tr…

Representation Learning Time Series

KAIROS: Unified Training for Universal Non-Autoregressive Time Series Forecasting

2025-10-02 · Kuiye Ding, Fanda Fan, Zheya Wang, Hongxiao Li 외 arxiv

In the World Wide Web, reliable time series forecasts provide the forward-looking signals that drive resource planning, cache placement, and anomaly response, enabling platforms to operate efficiently as user behavior an…

Zero-shot Generalization Time Series Forecasting Decision Making

DualWeaver: Synergistic Feature Weaving Surrogates for Multivariate Forecasting with Univariate Time Series Foundation Models

2026-02-25 · Jinpeng Li, Zhongyi Pei, Huaze Xue, Bojian Zheng 외 arxiv

Time-series foundation models (TSFMs) have achieved strong univariate forecasting through large-scale pre-training, yet effectively extending this success to multivariate forecasting remains challenging. To address this,…

TimeSynth: A Framework for Uncovering Systematic Biases in Time Series Forecasting

2026-02-11 · Md Rakibul Haque, Vishwa Goudar, Shireen Elhabian, Warren Woodrich Pettine arxiv

Time series forecasting is a fundamental tool with wide ranging applications, yet recent debates question whether complex nonlinear architectures truly outperform simple linear models. Prior claims of dominance of the li…

Time Series Forecasting

In-Context Fine-Tuning for Time-Series Foundation Models

2024-10-31 · Abhimanyu Das, Matthew Faw, Rajat Sen, Yichen Zhou

Motivated by the recent success of time-series foundation models for zero-shot forecasting, we present a methodology for $\textit{in-context fine-tuning}$ of a time-series foundation model. In particular, we design a pre…

Time Series