Multi-Factor Trend and Momentum Signals from Moving Averages
Summary
This strategy combines trend and momentum readings into a threshold-based directional signal. It derives several moving averages at different horizons, compares their levels, and uses changes in those averages as momentum inputs. Sine and cosine transformations also contribute to a confluence score. The resulting factors are combined into a signed state, with configurable upper and lower thresholds used to determine long, short, or neutral positioning; a reverse-trading option can invert direction.
The document describes a BTC/USDT futures backtest configuration using daily bars from November 2022 to November 2023, but gives no performance results or statistical evidence for the approach. It highlights the challenge of selecting parameters, potential high turnover, and sensitivity to market behavior. The strategy description suggests adding stop logic, refining parameters, and using machine learning for market-state assessment, but does not establish that these changes improve performance. Its many interacting calculations and thresholds make validation important before relying on the signals.
Key ideas
- Multiple moving averages and their differences represent trend structure across horizons.
- Average changes, transformed trend terms, and momentum components contribute to a combined score.
- Thresholds map the score to long, short, or neutral states, with an optional direction reversal.
- The published BTC/USDT futures setup provides dates and bar frequency but no performance statistics.
- Parameter selection, possible high trading frequency, and market sensitivity are key limitations.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.