전략 유형
머신러닝 트레이딩 알고리즘
퀀트 금융을 위한 오픈소스 ML 모델
머신러닝 트레이딩 알고리즘은 통계 학습 모델(그래디언트 부스팅, 신경망, 트랜스포머)을 금융 시계열 데이터에 적용해 가격 움직임을 예측하거나 거래 신호를 생성합니다. 이러한 오픈소스 구현은 회귀(가격 목표)와 분류(방향) 두 가지 정형화를 모두 다룹니다.
12개 알고리즘2개 라이브러리
알고리즘 네트워크
ML 알고리즘이 라이브러리에 걸쳐 어떻게 연결되는지
ML Algorithms
Freqtrade9 algos
Microsoft Qlib6 algos
LightGBMRegressorintermediate
LightGBMClassifierintermediate
XGBoostRegressorintermediate
XGBoostClassifierintermediate
PyTorchMLPRegressoradvanced
PyTorchTransformerRegressoradvanced
LGBModelintermediate
XGBModelintermediate
DNNModeladvanced
ALSTMadvanced
TFTModeladvanced
GATsadvanced
트레이딩 결정 파이프라인
ML 알고리즘이 트레이딩 시스템에서 어떻게 협력하는지
1
Market Data Input
Price, volume, indicators
OHLCV price data
Technical indicators
Feature engineering
2
ML Signal Generation
Model predictions
LightGBM/XGBoost regression
Transformer sequence modeling
MLP classification
3
Entry Decision
Signal threshold filtering
Confidence score > 0.6
Direction: Long/Short
4
Exit Strategy
Profit taking & stop loss
Take profit (fixed % or trailing)
Stop loss (ATR-based)
5
Risk Management
Position sizing & portfolio
Position sizing (Kelly criterion)
Portfolio rebalancing
알고리즘 비교
핵심 차원에서 ML 알고리즘 비교
알고리즘 비교 매트릭스열을 클릭하여 세부 정보 펼치기
| 항목 | LightGBMRegressorFreqtrade | LightGBMClassifierFreqtrade | XGBoostRegressorFreqtrade | XGBoostClassifierFreqtrade | PyTorchMLPRegressorFreqtrade | PyTorchTransformerRegressorFreqtrade | LGBModelQlib (Microsoft) | XGBModelQlib (Microsoft) | DNNModelQlib (Microsoft) | ALSTMQlib (Microsoft) | TFTModelQlib (Microsoft) | GATsQlib (Microsoft) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 복잡도 | ⭐⭐⭐intermediate | ⭐⭐⭐intermediate | ⭐⭐⭐intermediate | ⭐⭐⭐intermediate | ⭐⭐⭐⭐advanced | ⭐⭐⭐⭐advanced | ⭐⭐⭐intermediate | ⭐⭐⭐intermediate | ⭐⭐⭐⭐advanced | ⭐⭐⭐⭐advanced | ⭐⭐⭐⭐advanced | ⭐⭐⭐⭐advanced |
| 예측 유형 | 회귀 | 분류 | 회귀 | 분류 | 회귀 | 회귀 | 혼합 | 혼합 | 혼합 | 시퀀스 | 혼합 | 혼합 |
| 훈련 속도 | ⚡⚡⚡ | ⚡⚡⚡ | ⚡⚡⚡ | ⚡⚡⚡ | ⚡⚡ | ⚡ | ⚡⚡ | ⚡⚡ | ⚡⚡ | ⚡ | ⚡⚡ | ⚡⚡ |
| 정확도 | 📊📊📊 | 📊📊📊 | 📊📊📊 | 📊📊📊 | 📊📊📊 | 📊📊📊📊 | 📊📊 | 📊📊 | 📊📊 | 📊📊📊 | 📊📊 | 📊📊📊📊 |
| 최적 용도 | 테이블 데이터 | 테이블 데이터 | 테이블 데이터 | 테이블 데이터 | 비선형 패턴 | 시계열 패턴 | 범용 | 범용 | 범용 | 순차 데이터 | 범용 | 그래프 관계 |
Freqtrade
Gradient boosting regression model for price movement prediction using LightGBM.
소스:
freqtrade/freqai/prediction_models/LightGBMRegressor.pyLightGBM classification model for user-defined discrete labels.
소스:
freqtrade/freqai/prediction_models/LightGBMClassifier.pyXGBoost-based regression model for continuous value prediction.
소스:
freqtrade/freqai/prediction_models/XGBoostRegressor.pyXGBoost classification model for user-defined discrete labels.
소스:
freqtrade/freqai/prediction_models/XGBoostClassifier.pyMulti-layer perceptron neural network for regression-based price prediction.
소스:
freqtrade/freqai/prediction_models/PyTorchMLPRegressor.pyQlib (Microsoft)
LightGBM model for stock return prediction using technical and fundamental features.
주요 파라미터
| num_leaves | 31 | Maximum number of leaves |
| learning_rate | 0.1 | Boosting learning rate |
소스:
qlib/contrib/model/gbdt.pyDeep neural network for nonlinear feature extraction and return prediction.
주요 파라미터
| hidden_size | 256 | Hidden layer size |
소스:
qlib/contrib/model/pytorch_nn.pyAttention-based LSTM for sequential stock data modeling with attention mechanism.
주요 파라미터
| hidden_size | 64 | LSTM hidden size |
| num_layers | 2 | Number of LSTM layers |
소스:
qlib/contrib/model/pytorch_alstm.pyTemporal Fusion Transformer combining static and temporal features for multi-horizon prediction.
소스:
qlib/contrib/model/pytorch_tft.py머신러닝 트레이딩 알고리즘, 알고리즘 참조
- LightGBMRegressor (Freqtrade)
- Gradient boosting regression model for price movement prediction using LightGBM. 소스: https://github.com/freqtrade/freqtrade/blob/c0a991f5632cb900075ff2864b6671ed71a319d7/freqtrade/freqai/prediction_models/LightGBMRegressor.py.
- LightGBMClassifier (Freqtrade)
- LightGBM classification model for user-defined discrete labels. 소스: https://github.com/freqtrade/freqtrade/blob/c0a991f5632cb900075ff2864b6671ed71a319d7/freqtrade/freqai/prediction_models/LightGBMClassifier.py.
- XGBoostRegressor (Freqtrade)
- XGBoost-based regression model for continuous value prediction. 소스: https://github.com/freqtrade/freqtrade/blob/c0a991f5632cb900075ff2864b6671ed71a319d7/freqtrade/freqai/prediction_models/XGBoostRegressor.py.
- XGBoostClassifier (Freqtrade)
- XGBoost classification model for user-defined discrete labels. 소스: https://github.com/freqtrade/freqtrade/blob/c0a991f5632cb900075ff2864b6671ed71a319d7/freqtrade/freqai/prediction_models/XGBoostClassifier.py.
- PyTorchMLPRegressor (Freqtrade)
- Multi-layer perceptron neural network for regression-based price prediction. 소스: https://github.com/freqtrade/freqtrade/blob/c0a991f5632cb900075ff2864b6671ed71a319d7/freqtrade/freqai/prediction_models/PyTorchMLPRegressor.py.
- PyTorchTransformerRegressor (Freqtrade)
- Transformer architecture for time-series regression using self-attention mechanism. 소스: https://github.com/freqtrade/freqtrade/blob/c0a991f5632cb900075ff2864b6671ed71a319d7/freqtrade/freqai/prediction_models/PyTorchTransformerRegressor.py.
- LGBModel (Qlib (Microsoft))
- LightGBM model for stock return prediction using technical and fundamental features. 주요 파라미터: num_leaves (Maximum number of leaves), learning_rate (Boosting learning rate).소스: https://github.com/microsoft/qlib/blob/main/qlib/contrib/model/gbdt.py.
- XGBModel (Qlib (Microsoft))
- XGBoost model for stock return prediction. 소스: https://github.com/microsoft/qlib/blob/main/qlib/contrib/model/xgboost.py.
- DNNModel (Qlib (Microsoft))
- Deep neural network for nonlinear feature extraction and return prediction. 주요 파라미터: hidden_size (Hidden layer size).소스: https://github.com/microsoft/qlib/blob/main/qlib/contrib/model/pytorch_nn.py.
- ALSTM (Qlib (Microsoft))
- Attention-based LSTM for sequential stock data modeling with attention mechanism. 주요 파라미터: hidden_size (LSTM hidden size), num_layers (Number of LSTM layers).소스: https://github.com/microsoft/qlib/blob/main/qlib/contrib/model/pytorch_alstm.py.
- TFTModel (Qlib (Microsoft))
- Temporal Fusion Transformer combining static and temporal features for multi-horizon prediction. 소스: https://github.com/microsoft/qlib/blob/main/qlib/contrib/model/pytorch_tft.py.
- GATs (Qlib (Microsoft))
- Graph Attention Networks modeling stock relationships for cross-stock prediction. 소스: https://github.com/microsoft/qlib/blob/main/qlib/contrib/model/pytorch_gats.py.