machine-learning-strategy

Community

Predict returns with walk-forward ML signals.

Authorloanntc
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill turns OHLCV market data into future-direction trading signals by training machine-learning models without look-ahead leakage.

Core Features & Use Cases

  • Walk-forward training: Uses expanding or sliding windows so each prediction is made using only historical data.
  • OHLCV feature engineering: Builds robust factors (momentum, volatility, RSI, moving-average ratios, volume ratios, Bollinger position, and intraday ratios) with sanitization and zero-division guards.
  • Signal generation contract: Produces clean outputs by mapping model probabilities to a continuous signal in [-1.0, 1.0], with NaN-safe fallbacks.
  • Use Case: When you have daily OHLCV for multiple symbols, you can generate per-symbol long/short confidence signals suited to a prediction horizon like N-day returns.

Quick Start

Use the machine-learning-strategy Skill to generate per-symbol signals from your OHLCV DataFrames by ensuring the columns open, high, low, close, and volume are present.

Dependency Matrix

Required Modules

scikit-learnpandasnumpy

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: machine-learning-strategy
Download link: https://github.com/loanntc/Paave/archive/main.zip#machine-learning-strategy

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.