JMA Trend Filtering with Reversal-Based Position Entries

Summary

The document presents a strategy that uses a family of JMA-style moving averages to classify trend direction and trade around reversals. Its description calls the approach a grid system: it proposes building short exposure in an uptrend and long exposure in a downtrend, with an optional candle-color filter. The source code, however, compares the shortest and longest averages to set the trend and submits long entries when that trend is up and short entries when it is down. It does not show grid spacing or repeated grid orders, so the written description and implementation differ.

The source defines averages from 10 to 200 periods and uses the shortest-versus-longest relationship for its trend state. Published backtest settings specify BTC/USDT futures on daily bars across roughly a year, but no results are provided. The document flags whipsaw, missed signals from candle filtering, and costs or position buildup from grid spacing as risks. Those grid-specific cautions are not demonstrated by the supplied code; risk controls and exits are also not fully detailed there.

Key ideas

• The source uses the relationship between 10-period and 200-period JMA-style averages to identify trend direction.
• The written strategy proposes opening grid positions against the prevailing trend at reversals.
• The supplied code instead enters in the direction of its calculated trend and does not implement visible grid spacing.
• A candle-color filter can restrict entries, while the published default disables it.
• Backtest settings are given, but the document reports no performance results or fully specified exit rules.

Tags

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