skills.rest
charliedream1
charliedream1COMMUNITY
Explore all skills by this author →
GitHub

ml-strategy

Generate walk-forward ML trading signals from OHLCV data.

6.1k|1.2k|Updated Jun 9, 2022
Search Tags:#backtesting#ml#ohlcv#machine-learning#feature-engineering#trading-signals#walk-forward
One-click install
npx skills add https://github.com/charliedream1/ai_quant_trade --skill ml-strategy-charliedream1
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ml-strategy
Source: https://github.com/charliedream1/ai_quant_trade/tree/main/a_%E5%85%A8%E7%BD%91%E4%BC%98%E7%A7%80%E8%B5%84%E6%BA%90/10_%E5%A4%A7%E6%A8%A1%E5%9E%8B/07_skill%E5%8C%85/vibe_trading_skills/ml-strategy
Command: npx skills add https://github.com/charliedream1/ai_quant_trade --skill ml-strategy-charliedream1

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill provides a framework to forecast future price directions using ML on OHLCV data and to generate robust trading signals while guarding against data leakage and overfitting.

Core Features & Use Cases

  • Walk-forward training to avoid look-ahead bias in financial time series.
  • Feature engineering to derive momentum, volatility, RSI, and other factors from OHLCV data.
  • Signal generation converting model probabilities into continuous [-1, 1] signals for backtesting and live trading.
  • Supports multiple models (RandomForest, GradientBoosting, Ridge) for flexible experimentation.
  • Applicability to equities, futures, crypto with data-quality checks.

Quick Start

Provide an OHLCV dataset to generate walk-forward trading signals.

Frequently Asked Questions about ml-strategy

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I use machine learning to generate trading signals from OHLCV data?▼

Walk-forward training prevents data leakage and look-ahead bias in financial time series by sequentially training the model on past data and testing on subsequent periods, ensuring robust out-of-sample trading signals.

Can I use scikit-learn models to backtest crypto trading strategies?▼

Yes, scikit-learn, pandas, and numpy are required to train models and generate signals. You provide an OHLCV dataset to start the walk-forward training and output continuous trading signals for backtesting.

What is walk-forward training in machine learning for financial time series?▼

Walk-forward training prevents data leakage and look-ahead bias in financial time series by sequentially training the machine learning model on past data and testing on subsequent periods, ensuring robust out-of-sample trading signals.

Can I use scikit-learn models to backtest crypto trading strategies?▼

Yes, you can use scikit-learn models like RandomForest, GradientBoosting, and Ridge to backtest crypto trading strategies, as the framework supports equities, futures, and crypto with modular feature engineering.

What's the best way to avoid overfitting when building ML-based trading signals?▼

The best way to avoid overfitting when building ML-based trading signals is to use walk-forward training, which prevents look-ahead bias and data leakage by continuously validating the model on unseen future data.

Do I need Python and pandas to generate trading signals from OHLCV data?▼

Yes, Python with scikit-learn, pandas, and numpy are required to train models and generate signals. You provide an OHLCV dataset to start the walk-forward training and output continuous trading signals for backtesting.

Related Skills

Explore More Data & Analytics Skills→
DaddyElonMusk69DaddyElonMusk69

ml-strategy

Generate directional trading signals from OHLCV data using walk-forward machine-learning training.

Community
Advanced
robonet-techrobonet-tech
1

improve-trading-strategies

Refine, optimize, and enhance trading strategies with Allora signals.

Official
Advanced
FrancyJGLisboaFrancyJGLisboa
2.2k

stock-analyzer-cskill

Generate technical stock analysis and trading signals from RSI, MACD, and Bollinger calculations.

Community
Intermediate
20YN0420YN04

ml-strategy

Generate predictive trading signals from OHLCV data using machine learning models.

Community
Advanced