Selecting Equity Factors and Building Portfolios with Incremental Explanatory Power

Summary

This report describes a process for choosing equity factors and turning them into portfolios. It first narrows a larger set of candidate factors by weighing their correlations, volatility, cumulative returns, Sharpe ratios, and economic rationale. It then selects a smaller group based on how much each factor adds to the model’s overall explanatory power. The reported candidates include A-share free-float market capitalization, book-to-price, and price momentum or reversal measures.

To distinguish factor contributions to excess returns, the report uses a layered incremental-explanation approach. It also presents three portfolio construction methods, focusing on a pure-factor method and quadratic optimization, with an example using CSI 500 constituents. The summary reports that performance measures can move in opposing directions, so factor selection involves trade-offs. The document is only an abstract and does not provide the underlying tests, detailed implementation, or evidence needed to assess robustness; one listed factor description is also incomplete.

Key ideas

• Factor selection weighs correlation, volatility, return measures, Sharpe ratios, and economic rationale together.
• The report narrows candidates by their incremental contribution to overall model explanatory power.
• A layered analysis is used to separate factors’ explanatory contributions to excess returns.
• The portfolio methods include pure-factor construction and quadratic optimization.
• The example uses stocks in the CSI 500 universe.

Tags

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