Improving Factor Crowding Measures with Pair Correlation and Volatility
Summary
This report summary examines ways to improve measures used to monitor crowding in Chinese equity factors. It focuses on pair correlation and factor volatility, two of four measure families previously tested alongside valuation spreads and long-term return reversal. The proposed revisions compare calculation approaches, including computing pair correlation from specific rather than total returns, and assess signals across original and orthogonalized factor sets.
The summary reports that revised pair-correlation measures can relate positively to future factor returns for some factors, with stronger and more widespread relationships when specific returns are used. They also appear useful for forecasting future return volatility, with further improvement from specific returns; results remain broadly similar after orthogonalization. Volatility based on the long side relative to the market has some predictive value but is generally weaker than a long-short measure, and its volatility forecasting weakens after orthogonalization. The long-short-to-market measure is described as weak. These are summary claims without sample details, statistical estimates, or the underlying report's full evidence.
Key ideas
- The report compares revised crowding indicators based on pair correlation and factor volatility.
- Pair correlation calculated from specific returns shows stronger reported links to future factor returns.
- Revised pair-correlation measures are reported to help forecast future factor return volatility.
- Pair-correlation results are broadly stable across original and orthogonalized factor sets.
- Long-side-to-market volatility has some predictive value, while the long-short-to-market measure is reported as weak.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.