Exp_AbsolutelyNoLagLwma_Digit_NN3_MMRec
This EA uses an Absolutely No Lag LWMA (Linear Weighted Moving Average) Digit indicator combined with a 3-layer Neural Network (NN3) for signal generation, enhanced with Money Management Recommendation (MMRec) for position sizing. It identifies trend direction through the no-lag LWMA and feeds digit patterns into a neural network to classify trade signals, then applies money management rules to calculate optimal lot sizes.
Open the interactive view (equity curve, symbol switch)Key facts
Backtest results (H1, about 1 year)
| Symbol | P&L | Annual | Max DD | Trades | Win rate | Sharpe | Recovery | Losing streak | Per trade | Status |
|---|---|---|---|---|---|---|---|---|---|---|
| EURUSD | 0.00% | 0.00% | 0.00% | 0 | — | — | — | — | — | Not convertible yet |
| GBPUSD | 0.00% | 0.00% | 0.00% | 0 | — | — | — | — | — | Not convertible yet |
| USDJPY | 0.00% | 0.00% | 0.00% | 0 | — | — | — | — | — | Not convertible yet |
| XAUUSD | 0.00% | 0.00% | 0.00% | 0 | — | — | — | — | — | Not convertible yet |
Why this EA never traded: an indicator ended up with a period of 0.
Converted from MQL5 to Python and run with the same rules for every EA. Positions use MT5 lot sizes (100,000 units per FX lot, 100 oz per gold lot) with 1:100 leverage, one order at a time, and a stop-out below 10% of the starting balance. Sharpe is computed from the equity curve. Results marked as unreliable are shown for completeness only.
Blow-up risk: No blow-up pattern found
None of the usual blow-up patterns were found in the code, and the account was not stopped out in our backtests. That is not a guarantee: any EA can lose money.
High: stopped out in our backtest, or martingale/grid sizing without a stop loss. Elevated: martingale or grid sizing with a stop loss, or several positions at once without a stop loss. What the label means and what we found →
Strategy analysis
The strategy concept is based on the Absolutely No Lag LWMA which eliminates the lag inherent in traditional moving averages, converting its values into digit patterns that are fed into a 3-layer feedforward neural network for signal classification. Entry conditions: long positions when the neural network outputs a buy signal, short positions on sell signals, with the LWMA direction serving as a trend filter confirmation. Exit conditions: positions are closed on reverse neural network signals or when stop loss/take profit targets are reached. Position sizing is dynamically calculated through the MMRec module based on account balance, risk percentage, and stop loss distance. Money management employs a fixed fractional model where each trade risks a pre-set percentage of account equity, with position size determined by the ratio of dollar risk to stop loss distance in points. The strategy combines trend-following with pattern recognition advantages, where the neural network learns the optimal mapping from digit patterns to trading signals through historical data training.
Strengths
- Combines no-lag LWMA with neural network for adaptive signal generation with strong pattern recognition capability
- Money management module dynamically calculates lot size based on risk percentage, effectively controlling per-trade risk
- No-lag moving average reduces the lag defect of traditional MA indicators, providing more timely signals
- Complete stop loss and take profit settings with clear risk boundaries
- Spread filter mechanism avoids entering trades in high-spread environments
Weaknesses
- Neural network weights require pre-training; performance may severely degrade when market regime differs from training data distribution
- No-lag LWMA, while reducing lag, can still generate frequent false signals in ranging markets
- Lacks adaptive adjustment mechanism for market volatility (e.g., ATR), using fixed-point stop loss and take profit
- Three-layer neural network structure is relatively simple and may not capture complex non-linear market relationships
- No explicit new bar check mechanism mentioned, posing risk of duplicate entries on the same bar
Possible improvements
- Introduce ATR-based adaptive stop loss/take profit to replace fixed-point settings, making risk management responsive to market volatility changes
- Add new bar check mechanism to prevent duplicate entries on the same bar, and include a ranging market filter (e.g., ADX) to reduce false signals
- Add online learning or periodic retraining mechanism for the neural network to adapt to changing market regimes
Risk assessment
Overall risk: Medium
- Fixed-point stop loss/take profit does not account for market volatility changes; stop loss may be too close during high volatility causing frequent stops
- Neural network may continuously generate wrong signals during market regime shifts, lacking regime detection and circuit breaker mechanisms
- Lack of explicit new bar check poses potential risk of duplicate entries on the same candle
Evaluation scores
Parameters
| Parameter | Default | Description |
|---|---|---|
Lots | 0.1 | Fixed lot size (when MM is disabled) |
MoneyManagement | true | Enable money management for automatic lot calculation |
RiskPercent | 2.0 | Risk percentage per trade (%) |
StopLoss | 500 | Stop loss in points |
TakeProfit | 1000 | Take profit in points |
TrailingStop | 0 | Trailing stop distance in points from current price |
NoLagLwma_Period | 14 | No-lag LWMA period |
NoLagLwma_Price | PRICE_CLOSE | Price type used for LWMA calculation |
NN_Layer1_Neurons | 32 | Neural network layer 1 neuron count |
NN_Layer2_Neurons | 16 | Neural network layer 2 neuron count |
NN_Layer3_Neurons | 2 | Neural network output layer neuron count |
DigitPeriod | 10 | Digital pattern extraction period |
MaxSpread | 30 | Maximum allowed spread |
MagicNumber | 123456 | EA magic number; 0 manages all positions |
Similar Trend Following EAs
Backtests are historical simulations and do not guarantee future results. Not financial advice.