Newest papers

STOCK-JEPA: Prior-Anchored Latent Revision Representation Learning in Equity Markets

Yizhi Luo, Jiahe Yi, Jianhui Zhang, Shuo Sun · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.07006 · cited by 0 works (OpenAlex) · open-access PDF

Korean

이 효과적인 컨텍스트 표현을 배우는 것은 낮은 신호-소음 비율을 가진 금융 데이터에서 신호 시장의 구조와 역학을 예측하는 데 도움이됩니다. 블랙박스 깊은 모델은 복잡한 패턴을 캡처 할 수 있지만 동일한 샘플 소음과 명시적 경제 구조가 부족 할 수 있습니다. 한편, 고전적인 선형 금융 모델은 해석 가능한 참조를 제공하지만, 그들의 과소 단순화 된 가정은 비선형 신호를 포착하지 않습니다.이 두 방향의 강점을 결합하기 위해 우리는 Stock-JEPA를 제안합니다. 동일한 삽입 된 예측 프레임 워크는 포인트-인-타임 금융 프리미엄에 대한 예측 가능한 연속적인 개정을 배웁니다. 첫째, 우리는 최적화 된 비선형

Machine translation (Meta M2M100); the English original follows.

Abstract

Learning effective representations helps characterize the structure and dynamics of equity markets from financial data with a low signal-to-noise ratio. Black-box deep models can capture complex patterns but may overfit sample noise and lack explicit economic structure. Meanwhile, classic linear financial models provide interpretable references, but their oversimplified assumptions leave non-linear signals uncaptured. To combine the strengths of these two directions, we propose Stock-JEPA, a joint-embedding predictive framework that learns predictable incremental revisions relative to a point-in-time financial prior. First, we leverage a low-complexity financial model to produce fixed statistics summarizing multi-horizon return and risk. A prior projector then maps these statistics into the target encoder's latent space as an anchor. Second, we design a context-conditioned revision predictor to estimate the future representation's predictable displacement from the anchor. Separate losses update the two branches: the anchor learns from prior statistics, while the revision captures additional predictable information from historical context. Third, we freeze all representation modules and train a downstream readout, evaluating its forecasts through cross-sectional ranking and portfolio performance. Theoretically, we prove that optimal revision reduces the prior anchor's expected squared error for the same future representation by exactly $\mathbb{E}[\|\boldsymbolΔ\|_2^2]$. This non-negative gain is the expected squared magnitude of the additional signal predictable from historical context. Experimentally, Stock-JEPA outperforms 13 strong baselines across large-scale China and U.S. equity universes on 5 key evaluation metrics. Ablation studies and representation analysis further demonstrate the value of the learned revisions for representation learning in equity markets.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Overfitting, Econometrics, Computer science, Leverage (statistics), Representation (politics), Equity (law)

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange