Exp_ColorJFatl_Digit_NN3_MMRec
This EA combines the ColorJFatl digital filter trend indicator with a 3-layer neural network (NN3) for signal confirmation, and includes a money management recommendation module for dynamic position sizing. It generates trade signals when the neural network confirms the digital filter's trend direction, applying recommended lot sizes based on account equity and risk parameters.
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: no entry rule survived the conversion.
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 core is based on the ColorJFatl digital filter indicator, which uses the FATL (Fast Adaptive Trend Line) digital filter to identify trend direction and turning points. Entry conditions require a ColorJFatl color change (trend reversal) confirmed by a 3-layer neural network (NN3), where the NN filters false signals by learning from historical price patterns — trades are only opened when the NN output aligns with the digital filter signal. Exit conditions include opposing signal appearance, stop-loss or take-profit triggers, and position closure when the money management recommendation module advises reducing exposure. Position sizing is controlled by the MMRec module, which dynamically calculates optimal lot sizes based on account balance, risk percentage, and stop-loss distance. The money management employs a recommendation mode that adjusts risk exposure based on consecutive wins and losses, increasing position size during winning streaks and reducing it during losing streaks.
Strengths
- Combines digital filter and neural network dual confirmation, effectively filtering false signals to reduce losing trades
- Built-in money management recommendation module dynamically adjusts positions based on profit/loss state, providing adaptive risk management
- The FATL digital filter effectively suppresses price noise, performing excellently in trend identification during smooth market conditions
Weaknesses
- The neural network requires sufficient historical data for training, making signal quality unreliable during cold-start phase
- The digital filter tends to produce frequent false reversal signals during ranging/choppy market conditions
- The 3-layer neural network architecture is relatively simple, with limited ability to capture complex non-linear market patterns
Possible improvements
- Add a ranging market identification filter (such as ADX or volatility indicator) to pause trading during consolidation periods and avoid false signals
- Introduce a deeper neural network architecture or LSTM structure to improve recognition of complex market patterns
Risk assessment
Overall risk: Medium
- Insufficiently trained neural network may produce incorrect confirmation signals leading to counter-trend position opening
- Digital filter lag increases during high volatility, potentially missing optimal entry timing or causing delayed stop-loss execution
Evaluation scores
Parameters
| Parameter | Default | Description |
|---|---|---|
SoundDuration | 5 | Alert sound duration in seconds |
MoneyRisk | 1.0 | Risk percentage per trade (%) |
StopLoss | 500 | Stop loss in points |
TakeProfit | 2000 | Take profit in points |
OrderComment | Order comment | |
TrailingStop | 0 | Trailing stop distance in points from current price |
Lots | 0.1 | Fixed lot size (when MM is disabled) |
MMMode | 1 | Money management mode (0=fixed lots, 1=recommended) |
JFatlPeriod | 12 | JFatl indicator period |
JFatlMethod | MODE_SMA | JFatl averaging method |
NN3Inputs | 3 | Neural network input layer nodes |
NN3Hidden | 7 | Neural network hidden layer nodes |
NN3Outputs | 1 | Neural network output layer node count |
Deviations | 1.0 | Signal deviation threshold |
Similar Machine Learning EAs
Backtests are historical simulations and do not guarantee future results. Not financial advice.