Measuring Stock Herding with Intraday Data Using the LSV Model

Summary

This research summary describes a stock-selection factor intended to measure herding in China’s A-share market. It applies the LSV model to intraday transaction records, classifying trades as buyer- or seller-initiated to estimate the relative strength of buying and selling in each stock. The resulting signal is evaluated across the broad market and several index constituent universes, with weekly portfolio rebalancing.

The summary reports that factor-ranked returns were differentiated and broadly monotonic in the tested universes. For a market-wide long portfolio hedged against the CSI 500, it gives backtest figures for information coefficient and annualized return, volatility, and information ratio; winsorization, neutralization, and standardization are reported to improve results. These are historical backtest claims summarized from a report, not guarantees of future performance. The summary does not supply the full methodology, implementation details, or transaction-cost analysis, and explicitly cautions that market conditions may change.

Key ideas

• The factor estimates stock-level herding from intraday transaction data classified by active buying and selling.
• The LSV model is used to quantify the degree of coordinated trading behavior.
• The summary reports return separation and monotonicity across several A-share stock universes.
• Weekly rebalanced long portfolios are evaluated, including a market-wide portfolio hedged against the CSI 500.
• Factor cleaning and normalization are reported to improve results, while future market changes remain a key limitation.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.