Opening-Price Shortcut in Stock Prediction: A Conditional Opening Prior-and-Evidence Framework

2026-10-03

AI Summary

This study addresses the problem in stock prediction where high correlation between opening and closing prices causes models to degenerate into an "opening price shortcut." To mitigate this, we propose the COPE framework, which explicitly models the opening state as a conditional variable and reformulates closing price prediction as a discrimination task of continuation or reversal relative to the opening direction. By leveraging a conditional prior to guide evidence decomposition and residual learning, COPE eliminates shortcut dependence on opening prices, extracting effective predictive signals solely from the price modality. Experiments on four real-world market datasets demonstrate that COPE significantly outperforms existing methods. Furthermore, backtesting validates a strong alignment between predictive accuracy and actual trading returns.

Abstract

Stock price prediction has long been a central problem in quantitative finance. While recent methods increasingly leverage external modalities such as news to enhance predictive performance, the microstructure within price sequences itself contains crucial information. The target-day opening price represents the first concentrated realization of overnight information in the market and holds significant value for intraday evolution and the eventual closing direction. However, once opening information is introduced, the opening and closing directions often exhibit high correlation, making the model prone to degenerating into an opening-price shortcut that simply extrapolates the closing direction from the opening direction. To address this, we propose \cope{} (\textbf{C}onditional \textbf{O}pening \textbf{P}rior and \textbf{E}vidence), a shortcut-aware prediction framework that relies solely on the price modality. \cope{} models the target-day opening state as an observed conditioning variable and reparameterizes the closing-direction prediction into a continuation/reversal discrimination conditioned on the opening direction. Furthermore, we decompose the input evidence into opening condition, individual historical state, and contextual support, and perform residual modeling of historical and contextual evidence under this condition to distill information truly effective for the final closing judgment. Systematic experiments on four real-market datasets demonstrate that \cope{} significantly outperforms existing methods. Ablation studies and shortcut-sensitive analyses further validate the effectiveness of explicitly modeling opening information, while opening-time backtests reveal that predictive accuracy and trading returns align.

Problem

• stock price prediction
• opening-price shortcut
• closing direction
• quantitative finance

Innovation

• Stock Prediction
• Shortcut Learning
• Conditional Prior
• Residual Modeling
• Market Microstructure