Time Series Foundation Models for Multivariate Financial Time Series Forecasting

Paper: arXiv 2507.07296

Authors: Ben A. Marconi

Abstract

Financial time series forecasting presents significant challenges due to complex nonlinear relationships, temporal dependencies, variable interdependencies and limited data availability, particularly for tasks involving low-frequency data, newly listed instruments, or emerging market assets. Time Series Foundation Models (TSFMs) offer a promising solution through pretraining on diverse time series corpora followed by task-specific adaptation. This study evaluates two TSFMs (Tiny Time Mixers (TTM) and Chronos) across three financial forecasting tasks: US 10-year Treasury yield changes, EUR/USD volatility, and equity spread prediction. Results demonstrate that TTM exhibits strong transferability. When fine-tuning both the pretrained version of TTM and an untrained model with the same architecture, the pretrained version achieved 25-50% better performance when fine-tuned on limited data and 15-30% improvements even when fine-tuned on lengthier datasets. Notably, TTM’s zero-shot performance outperformed naive benchmarks in volatility forecasting and equity spread prediction, with the latter demonstrating that TSFMs can surpass traditional benchmark models without fine-tuning. The pretrained model consistently required 3-10 fewer years of data to achieve comparable performance levels compared to the untrained model, demonstrating significant sample-efficiency gains. However, while TTM outperformed naive baselines, traditional specialised models matched or exceeded its performance in two of three tasks, suggesting TSFMs prioritise breadth over task-specific optimisation. These findings indicate that TSFMs, though still nascent, offer substantial promise for financial forecasting-particularly in noisy, data-constrained tasks-but achieving competitive performance likely requires domain-specific pretraining and architectural refinements tailored to financial time series characteristics.

Complexity vs Empirical Score

• Math Complexity: 3.5/10
• Empirical Rigor: 7.5/10
• Quadrant: Street Traders — practical and empirical, lighter on theory

Why this score: This paper presents a strong empirical evaluation of Time Series Foundation Models (TTM and Chronos) across diverse financial asset classes. While it relies primarily on applied deep learning architectures rather than complex theoretical mathematical derivations, its sample-efficiency testing and transferability analysis provide valuable empirical insights.

Research Flowchart

  flowchart TD
    A[Evaluate TSFMs for Multivariate Financial Forecasting] --> B[Input Data: US Treasury Yields, EUR/USD Volatility, Equity Spreads]
    B --> C[Evaluate TSFMs: TTM & Chronos]
    C --> D[Testing Modes: Zero-Shot Inference vs Task-Specific Fine-Tuning]
    D --> E[Benchmark Comparison against Traditional Specialized Models & Naive Baselines]
    E --> F[Findings: Pretrained TTM improved performance by 15-50% & required 3-10 years less data]
    E --> G[Outcome: TSFMs excel in data-constrained settings but need domain-specific pretraining for top performance]

Cited References (133)

1. Uniform Output Regulation of Nonlinear Systems: A Convergent Dynamics Approach (2006)
2. $\epsilon$-Approximation of differential inclusions (1996)
3. A New Approach to the Economic Analysis of Nonstationary Time Series and the Business Cycle (1989) — link
4. Regime-Switching Global Vector Autoregressive Models (2013) — link
5. Regime-Switching Factor Investing with Hidden Markov Models (2020) — link
6. A Machine Learning Approach to Regime Modeling (2021)
7. Clustering Market Regimes using the Wasserstein Distance (2021) — link
8. Machine Learning and the Yield Curve: Tree-Based Macroeconomic Regime Switching (2024) — link

2 of 133 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.