Skip to content
All library documents

Moving-Average Signal with Percentage Stop and Target Orders

Article Strategy library · Author: ChaoZhang

Summary

This document describes a simple automated strategy that pairs a three-period moving average with user-defined percentage stop-loss and take-profit levels. It says to enter long when the average crosses above zero and short when it crosses below zero, then set protective and profit-taking prices relative to the entry price. It also discusses canceling exit orders after a position closes and plotting the levels. Published backtest settings specify BTC/USDT futures, daily bars, and a one-hour base period from January to November 2023, but no performance results are provided.

The source does not fully support the prose description. It calculates the average from an average-price series, places a long limit order whenever the average is positive, and defines short orders as exits; it does not implement the stated short-on-cross-below-zero rule. The entry offset and percentage exit orders are configurable, but the source uses entry-order calls for the exits, so their behavior may differ from conventional attached stop and target orders. The document identifies false signals, unsuitable risk percentages, and drawdown as concerns, and suggests filters and position sizing. Its backtest settings alone do not establish effectiveness.

Key ideas

  • The described framework uses a three-period simple moving average and user-set percentage stop and target levels.
  • The prose describes entries when the average crosses zero, but the supplied source instead checks whether it is above zero.
  • The source defines a long entry offset and uses order calls for take-profit and stop orders, which may not behave like attached exits.
  • No performance statistics are reported, and the listed backtest settings alone do not demonstrate profitability.
  • The document flags signal quality, risk parameter choice, and drawdown as concerns.

Tags

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