.png){kind=icon}
.png){kind=icon}
Return
Understanding Interactive Stock Dynamics via Sensitivity-Aware Dependency Learning
.png){kind=icon}
PRE
AI
DOI：10.1109/TKDE.2026.3675433.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The inherent fluctuations in the stock market present significant challenges in understanding stock dynamics, especially for investment decisions based on stock ranking. Recent advancements in learning-based methods have led to promising results in exploring temporal dependencies to understand stock movements. However, they often assume stable, certain, and reliable environments, narrowing their insight into the complex and fluctuating nature of markets. This complexity is driven by two influential factors: the explicit consistency of dynamic yet stable trends across diverse temporal patterns, coupled with the implicit interplay of logic and possibility under uncertainty. Hence, we introduce a <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">S</b>ensitivity-aware <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">D</b>ependency <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">L</b>earning solution (<bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SDL</b>) for stock ranking. With bridging the ideal and reality in mind, SDL captures short-term fluctuations under the guidance of long-term dependencies, associated with the augmentation of counterfactual knowledge. Specifically, SDL devises a <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">S</b>hort-term <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">C</b>o-integration <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">D</b>etector (SCD) that concentrates on capturing time-varying correlations and immediate market reactions, in addition to multi-period attention. Furthermore, a <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">L</b>ong-term <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">C</b>o-movements <bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">T</b>racker (LCT) takes advantage of enduring industry relationships and incorporates counterfactual knowledge, allowing the model to generalize beyond observed patterns and identify diverse long-term trends. Comprehensive experiments on five real-world stock markets demonstrate that our proposed SDL outperforms several representative baselines.
Keywords:
Counterfactual augmentation
dependency learning
dynamic relation learning
stock ranking
uncertain condition
Journal
IF:
10.4
Papers:
6.7K
Citations:
3.2W
Organization
View more organizations