Triple SMI Signals for Short-Term Trading

Summary

This short-term strategy combines three Stochastic Momentum Index (SMI) series, using lookback lengths of 10, 20, and 5 with smoothing periods of 3. It describes SMI as a bounded oscillator that compares the close with the high-low range, and identifies possible turning points using overbought and oversold levels around 40 and -40. The intended approach is to require agreement across all three settings, with SMI and signal-line relationships also used to indicate direction.

The published example specifies a BTC/USDT futures backtest on an hourly period with 15-minute base data over about one month, but reports no performance results. The strategy discussion proposes testing parameter variants, adding volume or Bollinger Bands, and refining stop-loss rules. Its main limitations are lag from smoothing, potentially high transaction costs from frequent short-term trades, overfitting, and sensitivity to market regime changes. The source code’s actual entry conditions compare each SMI with its signal line; they do not check the overbought or oversold thresholds described in the prose, so the stated rationale and implementation differ.

Key ideas

• The strategy compares three SMI calculations using lookback lengths of 5, 10, and 20.
• It uses agreement among the oscillator and its signal lines to produce directional conditions.
• The written rationale describes overbought and oversold thresholds, but the source entry rules do not apply those thresholds.
• Smoothing can delay signals, and frequent trading may make transaction costs significant.
• The document proposes parameter testing and robustness checks but provides no backtest performance results.

Tags

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