Gaussian-Smoothed Detrended Oscillator Reversion Strategy
Summary
This strategy uses a detrended price oscillator to generate potential mean-reversion signals. It subtracts a lagged exponential moving average from the close, then smooths the result with an Arnaud Legoux moving average. The smoothed oscillator is compared with a lagged copy of itself and with a zero line to define entries and exits.
A long entry occurs when the oscillator crosses above its lag while below zero; the long is closed on a cross below the lag or zero line. A short entry uses the inverse crossover while the oscillator is above zero, with exits on the corresponding upward crosses. The chart plots the oscillator, lag, and zero reference, and highlights entries and exits. The document explains the signal construction but supplies no backtest metrics, market-specific settings, transaction-cost assumptions, or evidence that the signals predict reversals. Performance and sensitivity to the chosen lengths therefore remain unestablished.
Key ideas
- The oscillator subtracts a lagged EMA from closing price to detrend the series.
- An ALMA smooths the detrended values before signal comparisons are made.
- Long entries use an upward cross of the oscillator over its lag while the oscillator is negative.
- Short entries use a downward cross below the lag while the oscillator is positive.
- Positions close on a reverse lag crossover or a crossover of the zero line, but no performance evidence is provided.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.