zhutoutoutousan
605faf5310
Add cluster audit pipeline, united EA updates, brochure generators, and publication hygiene (gitignore, MT5 path desensitization, pre-upload scan). Remove tracked reports, models, and binary artifacts from the repo. Co-authored-by: Cursor <cursoragent@cursor.com>
135 lines
4.2 KiB
Markdown
135 lines
4.2 KiB
Markdown
# Genetic Algorithm Optimization Guide for RSIScalpingNVDA
|
|
|
|
## Recommended Optimization Strategy
|
|
|
|
### Phase 1: Core RSI Parameters (Primary Focus)
|
|
These parameters directly control entry/exit signals and should be optimized first.
|
|
|
|
#### **RSI_Period** (Y - Optimize)
|
|
- **Current**: 14
|
|
- **Recommended Range**: 7-21
|
|
- **Step**: 1
|
|
- **Rationale**: Standard RSI periods. Shorter = more sensitive, longer = smoother signals
|
|
|
|
#### **RSI_Overbought** (Y - Optimize)
|
|
- **Current**: 19.0 (unusually low - verify if this is correct)
|
|
- **Standard Range**: 60.0-85.0
|
|
- **Step**: 2.0
|
|
- **Alternative Range** (if current is intentional): 15.0-30.0
|
|
- **Rationale**: Level where RSI indicates overbought condition for sell entries
|
|
|
|
#### **RSI_Oversold** (Y - Optimize)
|
|
- **Current**: 50.0 (unusually high - verify if this is correct)
|
|
- **Standard Range**: 15.0-40.0
|
|
- **Step**: 2.0
|
|
- **Alternative Range** (if current is intentional): 40.0-60.0
|
|
- **Rationale**: Level where RSI indicates oversold condition for buy entries
|
|
|
|
#### **RSI_Target_Buy** (Y - Optimize)
|
|
- **Current**: 71.0
|
|
- **Recommended Range**: 65.0-90.0
|
|
- **Step**: 2.0
|
|
- **Rationale**: Exit target for long positions. Must be > RSI_Oversold
|
|
|
|
#### **RSI_Target_Sell** (Y - Optimize)
|
|
- **Current**: 70.0
|
|
- **Recommended Range**: 10.0-35.0
|
|
- **Step**: 2.0
|
|
- **Rationale**: Exit target for short positions. Must be < RSI_Overbought
|
|
|
|
### Phase 2: Risk Management Parameters
|
|
|
|
#### **BarsToWait** (Y - Optimize)
|
|
- **Current**: 1
|
|
- **Recommended Range**: 1-8
|
|
- **Step**: 1
|
|
- **Rationale**: Bars to wait before closing when RSI goes against position. Higher = more patience
|
|
|
|
#### **TimeFrame** (Y - Optimize)
|
|
- **Current**: 16387 (M5)
|
|
- **Recommended**: Test M1, M5, M15, H1
|
|
- **Values**:
|
|
- M1 = 16385
|
|
- M5 = 16387
|
|
- M15 = 16388
|
|
- H1 = 16390
|
|
- **Rationale**: Different timeframes can significantly affect scalping performance
|
|
|
|
### Phase 3: Position Sizing (Optimize with Caution)
|
|
|
|
#### **LotSize** (Y - Optimize with Fixed Risk)
|
|
- **Current**: 50.0
|
|
- **Recommended Range**: 10.0-100.0
|
|
- **Step**: 5.0
|
|
- **Note**: Consider using fixed risk % instead of fixed lot size
|
|
- **Rationale**: Position sizing affects profitability but also risk
|
|
|
|
### Fixed Parameters (Do NOT Optimize)
|
|
|
|
#### **RSI_Applied_Price** (N)
|
|
- **Value**: 1 (PRICE_CLOSE)
|
|
- **Rationale**: Standard choice, changing may not improve results significantly
|
|
|
|
#### **MagicNumber** (N)
|
|
- **Value**: 12345
|
|
- **Rationale**: Identifier only, no impact on performance
|
|
|
|
#### **Slippage** (N)
|
|
- **Value**: 3
|
|
- **Rationale**: Broker-specific, should match your actual slippage
|
|
|
|
## Genetic Algorithm Settings
|
|
|
|
### Recommended GA Settings:
|
|
- **Optimization Criterion**: Balance (or Custom: Profit Factor * Total Net Profit)
|
|
- **Population Size**: 50-100
|
|
- **Mutation Probability**: 0.1-0.2
|
|
- **Crossover Probability**: 0.7-0.9
|
|
- **Optimization Passes**: 3-5
|
|
- **Forward Testing**: Always use out-of-sample data
|
|
|
|
### Optimization Phases:
|
|
|
|
1. **Broad Search** (First Pass):
|
|
- Optimize: RSI_Period, RSI_Overbought, RSI_Oversold, RSI_Target_Buy, RSI_Target_Sell
|
|
- Fix: BarsToWait=1, TimeFrame=M5, LotSize=50
|
|
|
|
2. **Refinement** (Second Pass):
|
|
- Use best results from Phase 1
|
|
- Optimize: BarsToWait, TimeFrame
|
|
- Narrow ranges around Phase 1 winners
|
|
|
|
3. **Fine-Tuning** (Third Pass):
|
|
- Optimize: LotSize (if needed)
|
|
- Very narrow ranges around Phase 2 winners
|
|
|
|
## Important Notes
|
|
|
|
⚠️ **Current Parameter Anomaly**:
|
|
- RSI_Overbought=19 and RSI_Oversold=50 are unusual
|
|
- Standard RSI ranges: Overbought 70-80, Oversold 20-30
|
|
- **Verify** if these are intentional or if there's a scaling issue
|
|
|
|
✅ **Validation Checklist**:
|
|
- Ensure RSI_Target_Buy > RSI_Oversold
|
|
- Ensure RSI_Target_Sell < RSI_Overbought
|
|
- Test on sufficient historical data (at least 6-12 months)
|
|
- Use forward testing on unseen data
|
|
- Check for overfitting (too many parameters optimized)
|
|
|
|
## Example .set File Structure
|
|
|
|
```
|
|
RSI_Period=14||1||7||21||Y
|
|
RSI_Overbought=70.0||2.0||60.0||85.0||Y
|
|
RSI_Oversold=30.0||2.0||15.0||40.0||Y
|
|
RSI_Target_Buy=75.0||2.0||65.0||90.0||Y
|
|
RSI_Target_Sell=25.0||2.0||10.0||35.0||Y
|
|
BarsToWait=2||1||1||8||Y
|
|
TimeFrame=16387||0||16385||16390||Y
|
|
LotSize=50.0||5.0||10.0||100.0||Y
|
|
RSI_Applied_Price=1||0||1||1||N
|
|
MagicNumber=12345||0||12345||12345||N
|
|
Slippage=3||0||3||3||N
|
|
```
|