Every portfolio manager has a stop-loss policy — either an explicit one she designed, or an implicit one she will discover at the worst possible moment. The difference is not whether one exists but whether its parameters are calibrated rationally to her Sharpe ratio, investment horizon, and the fund's loss tolerance. An explicit stop-loss is also a partial hedge for the PM's own incentive problem: because the PM is paid a percentage of profits, she holds a call option on portfolio returns — an option whose value increases with volatility, giving her an incentive to run excessive risk at the fund's expense. Stop-loss rules constrain this option. Understanding the cost-benefit trade-off of stop-loss thresholds is a prerequisite for setting them correctly.
In this lesson
- Mechanics of a stop-loss
- Rationale for stop-loss policies
- The Grossman-Zhou optimal proportional policy
- Costs of stop-loss
- Efficiency analysis
Before you start
- Performance Attribution: Selection, Sizing, and Timing
- Sustainable Leverage: Sizing the Book for Long-Term Viability
The full lesson, with worked examples and exercise solutions, is for members.
Read the full lessonNew to SelfMadeQuant.com? See membership options or browse the Portfolio Theory and Quantitative Investment syllabus.