Evaluating a 20-Day Trading-Value Factor for Five-Day Stock Returns
Summary
This Chinese-language research note analyzes recent trading value as a liquidity factor for stock selection. It describes changing the evaluation horizon to five-day returns, excluding stocks at their daily upper price limit to reflect execution difficulty and possible factor behavior differences, and comparing portfolio-group returns with a bar chart. The author reports an information coefficient of 0.052 and a long-side turnover rate of 4.2%. The highest-return group is reported as the fifth group, with a bell-shaped pattern across groups and visible differences between them.
The note says long-side returns were negative in 2019 and 2022 and positive in other years considered, indicating weaker results in bear markets. It also reports that neutralization reduced monotonicity, which the author attributes to correlation with market capitalization. For linear combinations, the note suggests transforming the factor to improve monotonicity; it says tree-based models can use it directly. These are reported findings, not independently verifiable results: the document gives no sample definition, full backtest methodology, transaction-cost treatment, or uncertainty estimates, and the factor may be regime-sensitive.
Key ideas
- The analysis treats recent trading value as a liquidity factor and evaluates its relationship with five-day returns.
- The author reports an information coefficient of 0.052 and long-side turnover of 4.2%.
- Returns across factor groups are described as bell-shaped, with the fifth group highest.
- Long-side results were negative in 2019 and 2022 and positive in the other years discussed.
- Market-cap neutralization reportedly weakened monotonicity, motivating factor transformation for linear combinations.
- The note provides limited methodology and no uncertainty or transaction-cost details.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.