Optimal Stopping and Bubble Riding under Limited Arbitrage Constraints: An Empirical Study based on A-Share Market Microstructure

Wensheng Yi, Zijian Zeng, Gao Ke, Jiaming Chen

Abstract

The Chinese A-share market, characterized by its unique retail-dominated structure and institutional constraints, exhibits pricing features distinct from mature markets. This study constructs a portfolio management framework adapted to the microstructure of the A-share market and examines how institutional investors can generate excess returns through Bubble Riding and Factor Timing strategies when limited arbitrage and investor irrationality coexist. First, using a synchronization-risk framework and a behavioral overreaction mechanism, we derive that the optimal strategy for rational arbitrageurs under heterogeneous beliefs and short-sale constraints can shift from immediate mean reversion to momentum-following. Subsequently, we integrate a China-localized factor construction and conduct an empirical study using A-share data from 2020 to 2026 collected through Python and the Baostock interface. The results show that: (1) traditional Fama-French factors are weakened in the A-share market when shell-value contamination is ignored, and excluding the bottom 30% of micro-cap stocks improves pricing efficiency; (2) momentum and sentiment factors exhibit nonlinear characteristics across market regimes; and (3) a dynamic risk-control strategy based on market-wide turnover rates—penalizing high-volatility exposure during bubble periods and increasing beta exposure during freezing periods—substantially outperforms the CSI 300 index out of sample.

Source: semanticscholar · PDF

Read the AI summary, key takeaways and discussion on WOBR Quant Research.