EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Information-Theoretic Causal Feature Selection via Markov Blanket Discovery for Stock Return Direction Prediction: Evidence from China

Jiamei Zhou iD, Hongxu Wu, Shaoze Li iD

Entropy · 29.07.2026 · 28 (8) · 847

DOI: 10.3390/e28080847 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Entropy, online erschienen am 29.07.2026. Verfasst von Jiamei Zhou und 2 weiteren Autor:innen. Beteiligte Einrichtung: College of Computer and Cyber Security, Fujian Normal University, Fuzhou 350117, China. Die Arbeit nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Stock markets are complex adaptive systems whose feature dependencies evolve over time, and identifying which signals genuinely drive returns is fundamentally a problem of information. Traditional feature selection methods and linear models rely on statistical correlations and overlook causality, which weakens their predictive performance in non-linear, evolving markets. This paper develops an information-theoretic approach to causal feature selection based on Markov Blanket discovery. We apply the Iterative Parent–Child-based search of Markov Blanket (IPCMB), whose conditional-independence tests are conditional mutual information measures, to recover the Markov Blanket of next-month returns—the minimal feature set that carries all Shannon information about the target. We then pair it with a Classification and Regression Tree (CART), whose impurity-based splitting admits an information-gain interpretation, to predict the direction of Chinese A-share returns non-linearly. Using data on 2760 stocks, IPCMB selects 13 causal features from 72 candidates, and CART forecasts whether the next month’s return is positive. The empirical results show an accuracy of 58.0%, 7.7 percentage points above the all-features CART benchmark, and a 13-month cumulative return 35.88% higher than that of the CSI 300 index. The findings indicate that selecting features according to their causal information content and combining them with interpretable tree-based prediction can support more reliable investment decisions in emerging markets.

Schlagwörter

Stichwörter der Autor:innen

Chinese stock marketIPCMB-CARTMarkov Blanketcausal feature selectionconditional mutual informationstock direction prediction

Autor:innen und Institutionen

  • College of Computer and Cyber Security, Fujian Normal University, Fuzhou 350117, China.
  • School of Finance and Accounting, Fujian Business University, Fuzhou 350012, China.
  • School of Economics and Management, University of Chinese Academy of Sciences, Beijing 100190, China.

ORCID-Profile: Jiamei Zhou, Shaoze Li

Zitieren

APA

Zhou, J., Wu, H., & Li, S. (2026). Information-Theoretic Causal Feature Selection via Markov Blanket Discovery for Stock Return Direction Prediction: Evidence from China. Entropy, 28(8), 847. https://doi.org/10.3390/e28080847

Vancouver

Zhou J, Wu H, Li S. Information-Theoretic Causal Feature Selection via Markov Blanket Discovery for Stock Return Direction Prediction: Evidence from China. Entropy. 2026;28(8):847. doi:10.3390/e28080847

BibTeX
@article{Zhou2026,
  author = {Zhou, Jiamei and Wu, Hongxu and Li, Shaoze},
  title = {Information-Theoretic Causal Feature Selection via Markov Blanket Discovery for Stock Return Direction Prediction: Evidence from China},
  journal = {Entropy},
  year = {2026},
  volume = {28},
  number = {8},
  pages = {847},
  doi = {10.3390/e28080847}
}
Kostenlos starten →