Please rotate your device to landscape mode to view the charts.

Machine Learning in the Chinese Stock Market

Journal: Journal of Financial Economics

Date: 20220801

Author: Leippold, Markus; Wang, Qian; Zhou, Wenyu

Abstract:
We add to the emerging literature on empirical asset pricing in the Chinese stock market by building and analyzing a comprehensive set of return prediction factors using various machine learning algorithms. Contrasting previous studies for the US market, liquidity emerges as the most important predictor, leading us to closely examine the impact of transaction costs. The retail investors' dominating presence positively affects short-term predictability, particularly for small stocks. Another feature that distinguishes the Chinese market from the US market is the high predictability of large stocks and state-owned enterprises over longer horizons. The out-of-sample performance remains economically significant after transaction costs.

Link: Google Scholar

Key Findings

Liquidity Dominates Predictability

Liquidity emerges as the most important predictor in the Chinese stock market, contrasting with US market where trend indicators dominate

Retail Investor Impact

Strong predictability for small stocks due to retail investors' dominance, with monthly R² up to 7.27% compared to 0.4% in US market

SOE Performance

State-owned enterprises show higher predictability over longer horizons, with annual Sharpe ratios up to 4.12 for neural network models

Model Performance Across Market Segments

  • Small stocks (Bottom 30%) show significantly higher predictability
  • SOE stocks demonstrate lower short-term but higher long-term predictability
  • Overall market predictability exceeds US market levels

Neural Network Portfolio Performance

  • Performance remains robust even after including transaction costs
  • Long-only strategy maintains significant economic value
  • Strategy performs well during 2015 market crash and 2020 pandemic

Variable Importance by Category

  • Liquidity measures are most influential predictors
  • Fundamental factors rank second in importance
  • Traditional momentum factors show less importance than in US market

Contribution and Implications

  • First comprehensive machine learning analysis of Chinese stock market predictability
  • Demonstrates unique characteristics of emerging markets dominated by retail investors
  • Provides evidence that machine learning methods can generate economically significant returns even under trading constraints
  • Shows importance of market structure and investor composition in determining predictability patterns

Data Sources

  • Performance Chart: Based on Table 1 showing out-of-sample predictive R² for different market segments
  • Sharpe Ratio Chart: Constructed using data from Table 10 showing portfolio performance including transaction costs
  • Variable Importance Chart: Derived from Figure 2 showing characteristic importance for all models