Using Return and Volume Moments to Build High-Frequency Stock Factors

Summary

This research summary tests stock selection factors built from return and trading-volume data, using moments from the first through fourth orders. It places these measures in a broader taxonomy: factors can use one data dimension, combine dimensions at one level, or combine multiple measures across dimensions. The proposed economic explanations divide unusual trading patterns into overreaction, which may later reverse, and abnormal trading activity associated with intense buying and selling and greater return uncertainty.

The reported all-share backtests show mixed results across return residual and volume-based factors. Several have positive information ratios and long-short Sharpe ratios, while the residual standard deviation factor has a negative information ratio. These results suggest potential cross-sectional selection value, but the summary does not provide sample dates, construction details, transaction costs, or robustness tests. Its performance figures should therefore be read as study-specific evidence rather than a guarantee of tradable returns.

Key ideas

• The study evaluates return and volume factors based on moments from the first through fourth orders.
• It distinguishes overreaction patterns from unusual trading activity as explanations for factor behavior.
• The tested factors show differing stock-selection results, including one residual volatility measure with a negative information ratio.
• The summary reports backtest performance but omits costs, sample-period details, and robustness evidence.

Tags

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