# AUDUSD Multi-timeframe trend filter 1m

This strategy implements a multi-timeframe trend filter approach for AUDUSD on 1-minute candles using RandomForestClassifier.

## Overview

- **Pair**: AUDUSD
- **Timeframe**: 1m
- **Model**: RandomForestClassifier with feature engineering focused on multi-timeframe trend analysis
- **Goal**: Trend following strategy that filters signals across multiple timeframes (short, medium, long-term)

## Features Engineered

1. **Returns**: 1, 3, 6, 12 period returns
2. **ATR-normalized price action**: Range percentage, body percentage, close position
3. **Multi-timeframe trend filters**:
   - Short-term: 5 EMA vs 13 EMA
   - Medium-term: 13 EMA vs 34 EMA  
   - Long-term: 34 EMA vs 89 EMA
   - Price position relative to each EMA (5, 13, 34, 89)
4. **Rolling volatility percentile** (from idea): Fast/slow volatility ratio, volatility percentile
5. **Distance from EMAs** (from idea): 20, 50, and 200 period EMAs
6. **Prior swing high and swing low distance** (from idea)
7. **Volume features**: Z-score and ratio to moving average

## Configuration

See `quant.config.json` for hyperparameters:
- `lookback`: 100 candles for prediction
- `horizon`: 1 candle forward for labeling
- `threshold`: 0.0004 (4 pips) for ATR-normalized breakout
- `min_confidence`: 0.50 minimum probability for signal generation

## Usage

This template follows the PyP Quant Mode contract:
```python
def train(data, config):
    return model, metrics

def predict(model, market_data, config):
    return {"signal": "UP|DOWN|HOLD", "confidence": 0.0, "metadata": {}}
```

## Disclaimer

Educational template only. Not financial advice. Past performance does not guarantee future results.