Using a Neural Network to Filter Stop-Out Risk in a Mean-Reversion Strategy
Summary
The article frames stop-outs as a problem of price moving against a correct directional view before eventually reversing. It argues that simply widening stops creates inconsistent risk, while waiting for confirmation may only delay a stop-out. Its baseline strategy uses prior-day highs and lows as reference levels and a moving average to identify entries in a mean-reversion approach: after a level is broken, it looks for a return toward the average. The system is described for EURUSD on an M30 chart, with historical testing using real-tick data.
A refined version uses a deep neural network to filter trades considered likely to cause large losses, while the article’s conclusion acknowledges that stop-outs cannot be eliminated. It reports that roughly 60% of positions were still losing in the discussed results and says the method filtered many large unprofitable trades. The excerpt does not provide enough detail to assess the network’s inputs, validation method, out-of-sample performance, or robustness across markets and periods; the stated results should therefore be treated as limited backtest evidence.
Key ideas
- The proposed baseline combines prior-day price extremes with a moving average to seek mean reversion after a breakout.
- The article favors a fixed stop size over widening stops in response to volatility.
- A deep neural network is introduced to filter trades judged prone to large losses.
- The reported results still include a high share of losing positions, so the method does not eliminate stop-outs.
- The available account does not establish out-of-sample robustness or broad applicability.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.