Testing Technical Alpha Factors in China A-Shares
Summary
The report evaluates 16 technical factors previously identified in research as significant and independent, testing their performance in China’s A-share market. Most fare poorly in the study. Five factors—DOWNILLIQ, UPILLIQ, NCSKEW, DUVOL, and IVmonthly—show stronger reported results, with the first two relating to downside and upside illiquidity and the next two capturing aspects of crash risk. The report notes that the paired illiquidity factors are negatively related in this market, while the two crash-risk measures are strongly positively related.
Fama–MacBeth regressions are used to assess whether the candidate factors add information beyond 12 existing factors; DOWNILLIQ, UPILLIQ, and IVmonthly retain incremental value in that analysis. A subsequent factor-selection exercise removes a short-history pair, drops the existing ILLIQ factor, replaces a one-month reversal factor with three-month reversal and DUVOL, and adds IVmonthly and EIVOL. These findings are specific to the historical sample and methodology. The report warns that models based on past data may lose effectiveness and that extreme market conditions can affect results.
Key ideas
- The study tests 16 previously identified technical factors in China’s A-share market.
- Five factors show stronger reported results, including measures of illiquidity, crash risk, and monthly idiosyncratic volatility.
- Fama–MacBeth regressions identify three of those factors as providing incremental information beyond existing factors.
- Factor pruning changes the selected reversal and illiquidity measures and adds volatility-related factors.
- Historical factor results may not persist, especially under extreme market conditions.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.