Paper: arXiv 2503.06928

Abstract

Despite the growing attention to time series forecasting in recent years, many studies have proposed various solutions to address the challenges encountered in time series prediction, aiming to improve forecasting performance. However, effectively applying these time series forecasting models to the field of financial asset pricing remains a challenging issue. There is still a need for a bridge to connect cutting-edge time series forecasting models with financial asset pricing. To bridge this gap, we have undertaken the following efforts: 1) We constructed three datasets from the financial domain; 2) We selected over ten time series forecasting models from recent studies and validated their performance in financial time series; 3) We developed new metrics, msIC and msIR, in addition to MSE and MAE, to showcase the time series correlation captured by the models; 4) We designed financial-specific tasks for these three datasets and assessed the practical performance and application potential of these forecasting models in important financial problems. We hope the developed new evaluation suite, FinTSBridge, can provide valuable insights into the effectiveness and robustness of advanced forecasting models in finanical domains.

Complexity vs Empirical Score

  • Math Complexity: 4.0/10
  • Empirical Rigor: 6.0/10 (capped from 8.5: transaction costs not modelled; see penalty flags)
  • Quadrant: Street Traders — practical and empirical, lighter on theory

Why this score: The paper’s core contribution is an empirical evaluation suite with curated datasets, specific financial tasks, and new metrics (msIC, msIR), demonstrating high practical rigor. While it references advanced time series models (e.g., transformers, decomposition methods), the paper itself is more focused on application and benchmarking rather than deriving new mathematical theorems.

Research Flowchart

  flowchart TD
  A["Research Goal: Bridge TS models with Financial Asset Pricing"] --> B{"Methodology"}
  B --> C["Data Creation<br>3 Financial Datasets"]
  B --> D["Model Selection<br>10+ TS Forecasting Models"]
  B --> E["Metric Development<br>msIC, msIR, MSE, MAE"]
  C --> F["Financial Task Design"]
  D --> F
  E --> F
  F --> G["Computational Process:<br>Benchmarking & Evaluation"]
  G --> H["Key Findings:<br>FinTSBridge Evaluation Suite<br>Model Insights & Practical Potential"]

Cited References (61)

  1. Dynamic connectedness between stock markets in the presence of the COVID-19 pandemic: does economic policy uncertainty matter? (2021)
  2. Indexing and stock market serial dependence around the world (2019)
  3. How do crises spread? Evidence from accessible and inaccessible stock indices (2006)
  4. Time-varying pattern causality inference in global stock markets (2021)
  5. Efficient capital markets: A review of theory and empirical work (1970)
  6. Stock market prices do not follow random walks: Evidence from a simple specification test (1988)
  7. Returns to Buying Winners and Selling Losers: Implications for (1993)
  8. Multifactor explanations of asset pricing anomalies (1996)
and 53 more
  1. The efficient market hypothesis and its critics (2003)
  2. Machine learning and the stock market (2023)
  3. Machine learning in the Chinese stock market (2022)
  4. Empirical asset pricing via machine learning (2020)
  5. Taming the factor zoo: A test of new factors (2020)
  6. Comovement (2005)
  7. TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis (2023) — link
  8. Timexer: Empowering transformers for time series forecasting with exogenous variables (2024) — link
  9. TimeMixer: Decomposable Multiscale Mixing for Time Series Forecasting (2024)
  10. A Time Series is Worth 64 Words: Long-term Forecasting with Transformers (2023)
  11. Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting (2021)
  12. Informer: Beyond efficient transformer for long sequence time-series forecasting (2021)
  13. Fedformer: Frequency enhanced decomposed transformer for long-term series forecasting (2022)
  14. Crossformer: Transformer Utilizing Cross-Dimension Dependency for Multivariate Time Series Forecasting (2023) — link
  15. iTransformer : Inverted Transformers Are Effective for Time Series Forecasting (2024) — link
  16. TSM ixer: An All- MLP Architecture for Time Series Forecast-ing (2023) — link
  17. N-BEATS: Neural basis expansion analysis for interpretable time series forecasting (2020) — link
  18. Are transformers effective for time series forecasting? (2023)
  19. Non-stationary transformers: Exploring the stationarity in time series forecasting (2022)
  20. Koopa: Learning non-stationary time series dynamics with koopman predictors (2023)
  21. Micn: Multi-scale local and global context modeling for long-term series forecasting (2023)
  22. Spectral temporal graph neural network for multivariate time-series forecasting (2020)
  23. Reversible instance normalization for accurate time-series forecasting against distribution shift (2021)
  24. Long-term Forecasting with Ti DE : Time-series Dense Encoder (2023)
  25. N-BEATS: Neural basis expansion analysis for interpretable time series forecasting (2019)
  26. Deep learning for forecasting stock returns in the cross-section (2018)
  27. Decision support from financial disclosures with deep neural networks and transfer learning (2017)
  28. Deep learning for event-driven stock prediction (2015)
  29. Mastering Stock Markets with Efficient Mixture of Diversified Trading Experts (2023) — link
  30. AlphaFin: Benchmarking Financial Analysis with Retrieval-Augmented Stock-Chain Framework (2024)
  31. A Survey of Large Language Models for Financial Applications: Progress, Prospects and Challenges — → 2406.11903
  32. MacroHFT: Memory Augmented Context-aware Reinforcement Learning On High Frequency Trading (2024)
  33. Understand the current status, trends and challenges of cryptocurrency derivatives (2020)
  34. Relations among Bitcoin Futures, Bitcoin Spot, Investor Attention, and Sentiment (2023)
  35. Attention is all you need (2017)
  36. PSformer: Parameter-efficient Transformer with Segment Attention for Time Series Forecasting (2024) — link
  37. Deep time series models: A comprehensive survey and benchmark (2024) — link
  38. Adaptive Normalization for Non-stationary Time Series Forecasting: A Temporal Slice Perspective (2023)
  39. LLMs and Foundational Models: Not (Yet) as Good as Hoped. (2024)
  40. Commentary: Can LLMs Provide Good Forecasts? (2024)
  41. Are Language Models Actually Useful for Time Series Forecasting? (2024) — link
  42. Fundamental limitations of foundational forecasting models: The need for multimodality and rigorous evaluation (2024) — link
  43. How to use security analysis to improve portfolio selection (1973)
  44. Active Portfolio Management: A Quantitative Approach for Producing Superior Returns and Controlling Risk (2000)
  45. A survey on machine learning models for financial time series forecasting (2022)
  46. Time-series forecasting with deep learning: a survey (2021)
  47. Machine learning advances for time series forecasting (2023)
  48. Forecasting stock market volatility with a large number of predictors: New evidence from the MS-MIDAS-LASSO model (2022)
  49. The risk and return of equity and credit index options (2024)
  50. Risk spillovers across geopolitical risk and global financial markets (2023)
  51. Advances in financial machine learning (2018)
  52. Automatic de-biased temporal-relational modeling for stock investment recommendation (2024)
  53. Data-driven stock forecasting models based on neural networks: A review (2024)

1 of 61 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.