Breaking the Trend: How to Avoid Cherry-Picked Signals

ArXiv ID: 2504.10914 “View on arXiv”

Authors: Unknown

Abstract

Our empirical results show an impressive fit with the pretty complex theoretical Sharpe formula of a trend-following strategy depending on the parameter of the signal, which was derived by by Grebenkov and Serror (2014). That empirical fit convinces us that a mean-reversion process with only one time scale is enough to model, in a pretty precise way, the reality of the trend-following mechanism at the average scale of CTAs and as a consequence, using only one simple EMA, appears optimal to capture the trend. As a consequence, using a complex basket of different complex indicators as signal, do not seem to be so rational or optimal and exposes to the risk of cherry-picking.

Keywords: trend-following strategy, mean-reversion process, exponential moving average (EMA), signal parameter, CTAs, Equities/Futures (Trend-following)

Complexity vs Empirical Score

• Math Complexity: 7.0/10
• Empirical Rigor: 8.0/10
• Quadrant: Holy Grail
• Why: The paper employs advanced stochastic process modeling to derive a theoretical Sharpe ratio (high math) and validates it against empirical CTA benchmark data with explicit parameter optimization, indicating backtest-ready methodology (high rigor).

  flowchart TD
    A["Research Goal<br>Quantify Optimal Trend Signal"] --> B["Methodology<br>Empirical Fit vs Theoretical Model"]
    B --> C["Input Data<br>Asset Prices & CTAs"]
    C --> D["Computational Process<br>EMA Signal vs Sharpe Formula"]
    D --> E{"Outcome Analysis"}
    E -- Good Fit --> F["Finding 1<br>Single-Scale Mean-Reversion Sufficient"]
    E -- Complex Signals --> G["Finding 2<br>Risk of Cherry-Picking"]
    F --> H["Conclusion<br>Optimal: Simple EMA Signal"]
    G --> H