Feedback Control for Stabilizing a Moving Average Trading System

Summary

The article applies feedback control and linear system identification to a simple moving-average strategy. The benchmark buys when price is above the average and sells when it is below, with fixed strategy parameters and ATR-based trade levels. The controller observes the system during backtesting and uses its behavior to regulate trading without changing those parameters. The stated aim is to reduce inefficient activity and improve stability rather than forecast prices directly.

The experiment uses two years of historical data, optimizing the moving-average period on the first half and evaluating the benchmark and controlled system on the second. The author reports that after an initial observation period, the controlled version reduced losses, increased net profit from a loss to a gain, traded less, and improved win rate and profit factor. These are results from a single described backtest, and the article’s claims about improved stability and profitability need independent testing across markets and periods; they do not establish durable performance.

Key ideas

• Feedback control regulates a strategy’s behavior based on observed system responses rather than directly predicting prices.
• The benchmark uses a moving-average rule with fixed parameters and ATR-based trade levels.
• The controller learns from backtest behavior while leaving the strategy parameters unchanged.
• The reported test compares an optimized benchmark with a controlled version on a later portion of the historical data.
• The reported gains come from one experiment and require validation across additional markets and periods.

Tags

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