Factor Exposure Caps and Optimized Portfolio Risk and Return

Summary

The document examines how factor exposure limits in a multi-factor portfolio optimization model affect active returns and risk. Raising the caps can increase excess return, drawdown, and tracking error, so the change in return relative to the change in risk determines whether the portfolio’s risk-adjusted result improves. It compares CSI 300 and CSI 500 portfolios, reporting a smaller return response for CSI 300 and a larger one for CSI 500.

The explanation emphasizes that realized factor exposures drive outcomes; portfolios may not reach their preset limits. Industry neutrality can constrain realized exposures more strongly in the concentrated CSI 300 universe than in CSI 500. With more effective factors, each factor contributes less and realized exposures tend to be lower, making return less sensitive to tighter caps. The proposed approach sets caps from the rolling average of realized exposures under a relatively loose limit. The summary claims this can work across different return forecasts, but provides no detailed performance figures or test design. It flags changing factor effectiveness, omitted risk factors, and optimizer failure as limitations.

Key ideas

• Higher factor exposure caps can raise both excess return and portfolio risk.
• Realized exposures, rather than preset limits, determine a portfolio’s factor-driven return contribution.
• Industry neutrality may restrict factor exposures more in CSI 300 than in CSI 500 portfolios.
• As the number of effective factors grows, lowering exposure caps may have less effect on excess returns.
• A rolling average of realized exposures under a loose cap can guide updated limits.

Tags

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