Paper: arXiv 2605.13407

Authors: Author One, Author Two

Abstract

We introduce a novel approach to modeling financial price dynamics using vector-quantized discrete latent factors. Our method achieves improved performance in predicting stock movements and volatility.

Complexity vs Empirical Score

  • Math Complexity: 7.5/10
  • Empirical Rigor: 6.5/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The method is based on ‘vector-quantized discrete latent factors,’ which implies advanced machine learning and mathematical modeling, elevating the math score. The mention of ‘improved performance in predicting stock movements and volatility’ suggests empirical validation, moving the rigor score to the high category.

Research Flowchart

  flowchart TD
    A[Research Goal:\nModel financial price dynamics] --> B[Data Input:\nHistorical equities price data]
    B --> C[Methodology:\nVector-Quantized Discrete Latent Factor Learning]
    C --> D[Process:\nTime series encoding & factor discretization]
    D --> E[Outcome 1:\nImproved stock movement prediction]
    D --> F[Outcome 2:\nImproved volatility prediction]

Cited References (85)

  1. Film: Visual reasoning with a general conditioning layer (2018)
  2. Are transformers effective for time series forecasting? (2023)
  3. Online clustered codebook (2023)
  4. FactorGCL: A Hypergraph-Based Factor Model with Temporal Residual Contrastive Learning for Stock Returns Prediction (2025)
  5. Rsap-dfm: Regime-shifting adaptive posterior dynamic factor model for stock returns prediction (2024)
  6. Accurate multivariate stock movement prediction via data-axis transformer with multi-level contexts (2021)
  7. Explainable stock price movement prediction using contrastive learning (2024)
  8. STORM: A Spatio-Temporal Factor Model Based on Dual Vector Quantized Variational Autoencoders for Financial Trading (2024) — link
and 77 more
  1. SimStock: Representation Model for Stock Similarities (2023)
  2. Matcc: A novel approach for robust stock price prediction incorporating market trends and cross-time correlations (2024)
  3. Factorvae: A probabilistic dynamic factor model based on variational autoencoder for predicting cross-sectional stock returns (2022)
  4. Finformer: A static-dynamic spatiotemporal framework for stock trend prediction (2023)
  5. FactorVQVAE: Discrete latent factor model via Vector Quantized Variational Autoencoder (2025)
  6. Outrageously large neural networks: The sparsely-gated mixture-of-experts layer (2017) — link
  7. Neural discrete representation learning (2017)
  8. Maskgit: Masked generative image transformer (2022)
  9. Taming transformers for high-resolution image synthesis (2021)
  10. Vector Quantized Time Series Generation with a Bidirectional Prior Model (2023) — link
  11. Bert: Pre-training of deep bidirectional transformers for language understanding (2018) — link
  12. Attention is all you need (2017)
  13. Auto-encoding variational bayes (2013) — link
  14. Autoencoder asset pricing models (2021)
  15. The cross-section of expected stock returns (1992)
  16. Comparing cross-section and time-series factor models (2020)
  17. Capital asset prices: A theory of market equilibrium under conditions of risk (1964)
  18. Xgboost: A scalable tree boosting system (2016)
  19. A dual-stage attention-based recurrent neural network for time series prediction (2017) — link
  20. The cross-section of expected stock returns (1992)
  21. Efficient capital markets (1970)
  22. Return, risk and arbitrage (1973)
  23. On persistence in mutual fund performance (1997)
  24. Deep risk model: a deep learning solution for mining latent risk factors to improve covariance matrix estimation (2021)
  25. Characteristics are covariances: A unified model of risk and return (2019)
  26. Variational inference with normalizing flows (2015)
  27. Portfolio analysis (1952)
  28. Accurate Stock Movement Prediction with Self-supervised Learning from Sparse Noisy Tweets (2022)
  29. Listening to chaotic whispers: A deep learning framework for news-oriented stock trend prediction (2018)
  30. Equity research report-driven investment strategy in Korea using binary classification on stock price direction (2021)
  31. CNNpred: CNN-based stock market prediction using a diverse set of variables (2019)
  32. Knowledge-driven stock trend prediction and explanation via temporal convolutional network (2019)
  33. Hierarchical Multi-Scale Gaussian Transformer for Stock Movement Prediction. (2020)
  34. Stock transaction prediction modeling and analysis based on LSTM (2018)
  35. Deep learning-based stock price prediction using LSTM and bi-directional LSTM model (2020)
  36. FinRL: A deep reinforcement learning library for automated stock trading in quantitative finance (2020) — link
  37. Straightening Out the Straight-Through Estimator: Overcoming Optimization Challenges in Vector Quantized Networks (2023) — link
  38. Conditional image generation with pixelcnn decoders (2016)
  39. Proximal policy optimization algorithms (2017) — link
  40. Continuous control with deep reinforcement learning (2015) — link
  41. Soft actor-critic algorithms and applications (2018) — link
  42. A multi-dynamic-factor model for stock returns (1992)
  43. Empirical evaluation of gated recurrent neural networks on sequence modeling (2014) — link
  44. Presidential address: Discount rates (2011)
  45. A five-factor asset pricing model (2015)
  46. HireVAE: An Online and Adaptive Factor Model Based on Hierarchical and Regime-Switch VAE — → 2306.02848
  47. Common risk factors in the returns on stocks and bonds (1993)
  48. Long short-term memory (1997)
  49. Temporal convolutional networks for action segmentation and detection (2017)
  50. Graph attention networks (2017) — link
  51. Qlib: An ai-oriented quantitative investment platform (2020) — link
  52. Momentum strategies (1996)
  53. Bimodal characteristic returns and predictability enhancement via machine learning (2022)
  54. Language models are unsupervised multitask learners (2019)
  55. Gaussian error linear units (gelus) (2016) — link
  56. MASTER: Market-Guided Stock Transformer for Stock Price Forecasting (2024)
  57. Stock market’s price movement prediction with LSTM neural networks (2017)
  58. A unified approach to interpreting model predictions (2017) — link
  59. A simple framework for contrastive learning of visual representations (2020)
  60. Pytorch: An imperative style, high-performance deep learning library (2019)
  61. Decoupled weight decay regularization (2017) — link
  62. Roformer: Enhanced transformer with rotary position embedding (2024)
  63. Can Machines’ Learn’Finance? (2020)
  64. Major issues in high-frequency financial data analysis: A survey of solutions (2025)
  65. Is there a replication crisis in finance? (2023)
  66. Visualizing data using t-SNE (2008)
  67. An introduction to the bootstrap (1994)
  68. Empirical Asset Pricing via Machine Learning (2020) — link
  69. Forest through the Trees: Building Cross-Sections of Stock Returns (2025) — link
  70. Regime Changes and Financial Markets (2012) — link
  71. Empirical asset pricing via machine learning (2020)
  72. itransformer: Inverted transformers are effective for time series forecasting (2023) — link
  73. Automatic de-biased temporal-relational modeling for stock investment recommendation (2024)
  74. Multi-Scale Temporal Neural Network for Stock Trend Prediction Enhanced by Temporal Hyepredge Learning (2025)
  75. Analyzing transaction graphs via motif-based graph representation learning for cryptocurrency price prediction (2025)
  76. Hypergraph neural networks to predict stock movements by exploring higher-order relationships (2025)
  77. Reversible instance normalization for accurate time-series forecasting against distribution shift (2021)

1 of 85 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.