Risk Estimation for Trade Sequences with Fixed and Trailing Stops
Summary
The article extends risk analysis for single-asset trading systems, focusing on fixed stop-loss and take-profit exits and fixed trailing stops. It discusses when deal outcomes can be treated as independent and identically distributed, and proposes alternatives when that assumption is doubtful: split a system into simpler components, analyze a simplified version, or assess results while adjusting for dependence through other methods. Its framework assumes prices are approximately random walks and emphasizes the limits that this places on conclusions about indicators and expected returns.
For fixed stop and take-profit exits, returns are modeled as two outcomes, a loss or a gain determined by the stop-to-target ratio. The article uses a binomial test under a zero-expected-return random-walk null to assess whether observed gains plausibly show an advantage, then discusses risk calculations for the specified exit distributions. It also warns that optimization can produce impressive-looking results by chance, illustrated by a simulated system with random trade directions. These methods depend on simplified distributional assumptions and should not be treated as universal risk estimates.
Key ideas
- Trade sequence risk analysis depends on whether outcomes are independent and identically distributed.
- System decomposition or simplified models can help when those assumptions fail.
- Fixed stop and target exits produce a two-outcome return model tied to the reward-to-risk ratio.
- A binomial test can compare observed profitable trades with a zero-expectancy random-walk baseline.
- Optimization can make random systems appear successful, so apparent performance needs statistical scrutiny.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.