Exp_XWAMI_NN3_MMRec
This EA is a trend-following expert advisor based on the XWAMI indicator combined with a 3-layer neural network (NN3) for signal confirmation and a money management recommendation module (MMRec). It enters trades when the neural network confirms trend direction signals from XWAMI and applies dynamic position sizing based on its money management recommendations.
Open the interactive view (equity curve, symbol switch)Key facts
Backtest results (H1, about 1 year)
Not backtested yet — this EA could not be converted automatically.
Strategy analysis
The core strategy of this EA is trend following based on the XWAMI indicator, which is an adaptive moving average-type indicator that dynamically adjusts sensitivity to market conditions. Entry conditions rely on the XWAMI indicator generating a trend direction signal, which is then secondarily confirmed by a 3-layer neural network (NN3); positions are only opened when the neural network output aligns with the XWAMI direction. Exit conditions likely include opposing XWAMI signals, neural network signal reversals, or fixed stop-loss and take-profit levels. Position sizing is implemented through the MMRec module, which dynamically calculates optimal lot sizes based on account equity, historical profit/loss, and market volatility. The money management module evaluates current risk exposure and recommends whether to increase, decrease, or maintain existing positions. The overall strategy combines the trend-judging capability of technical indicators with the adaptive learning ability of neural networks, aiming to filter out false signals and improve win rates.
Strengths
- Neural network secondary confirmation effectively filters false signals from XWAMI, improving entry quality
- MMRec money management module dynamically adjusts position sizing based on market conditions for more flexible risk control
- XWAMI adaptive indicator adjusts sensitivity based on market volatility, adapting to different market conditions
- Multi-layer neural network architecture provides nonlinear fitting capability to capture complex market patterns
- The design philosophy combining trend following with machine learning is forward-looking
Weaknesses
- Neural network weights require sufficient training and regular retraining; otherwise performance degrades significantly when market structure changes
- High overfitting risk — the 3-layer neural network may learn noise rather than genuine patterns with limited training data
- XWAMI indicator still generates frequent false trend signals in ranging markets; even with neural network filtering, frequent stop-losses may occur
- Complex parameter coupling between modules makes tuning difficult, unsuitable for inexperienced users
- Lack of dedicated handling for slippage and spread may cause significant deviation between actual execution and backtesting under extreme market conditions
Possible improvements
- Add a market regime identification module to automatically reduce position sizing or pause trading during ranging markets to minimize losses from false signals
- Implement online learning for the neural network so it can continuously adapt to market structure changes without manual retraining
- Add slippage and spread filter conditions to ensure trades are not executed when transaction costs are excessively high
Risk assessment
Overall risk: Medium
- Neural network may generate incorrect signals leading to consecutive losses when insufficiently trained or during sudden market structure shifts
- Lack of explicit fixed stop-loss mechanism (when set to 0, relying on dynamic stops) — extreme market conditions may cause losses exceeding expectations
- Complex parameter coupling between modules — misconfiguration may lead to overtrading or excessive position sizing
Evaluation scores
Parameters
| Parameter | Default | Description |
|---|---|---|
MMRec_Enable | true | Enable money management recommendation module |
NN3_Layer1_Neurons | 32 | Number of neurons in neural network layer 1 |
NN3_Layer2_Neurons | 16 | Number of neurons in neural network layer 2 |
NN3_Layer3_Neurons | 8 | Number of neurons in neural network layer 3 |
XWAMI_Period | 14 | XWAMI indicator calculation period |
StopLoss | 0 | Fixed stop loss in points, 0 to disable |
TakeProfit | 0 | Fixed take profit in points, 0 to disable |
RiskPercent | 2.0 | Risk percentage for dynamic position sizing |
LotSize | 0.1 | Fixed lot size, used when MMRec is disabled |
Timeframe | H1 | TII calculation timeframe, 0 for current period |
Similar Trend Following EAs
Backtests are historical simulations and do not guarantee future results. Not financial advice.