AI Pulse Data Science Projects¶

This directory contains data science projects and analyses that support AI Pulse articles.

Folder Structure¶

Each article has its own project folder with the following structure:

YYYY-MM-DD-article-title/
├── notebooks/          # Jupyter notebooks demonstrating analysis
├── data/              # Raw and processed data files
├── scripts/           # Python scripts and analysis code
├── outputs/           # Generated charts, tables, and results
└── analysis/          # Final analysis summaries and insights

Current Projects¶

2025-10-01-vision-weekly-stop-loss-optimization¶

Article: Vision Weekly Optimization: Stop-Loss Testing Results
Focus: Stop-loss vs stop-gain matrix analysis using 25 train/test splits
Key Finding: 3% stop-loss with no stop-gain optimal (22.1% improvement)

2025-10-02-vision-weekly-walk-forward¶

Article: Vision Weekly Three-Dimensional Trading Optimization: Walk-Forward Entry Strategies
Focus: Testing implementable walk-forward entry methods vs immediate entry
Key Finding: Even Williams Fractals underperform immediate entry + systematic exits

Notebook Standards¶

Each project should include: 1. Data Loading & Preparation - Import and clean source data 2. Methodology Explanation - Document approach and assumptions 3. Analysis Implementation - Execute core calculations 4. Results Visualization - Generate charts and tables 5. Key Insights Summary - Highlight main findings

Data Sources¶

Vision Research LLC weekly short recommendations (2024-2025)
Yahoo Finance historical price data via yfinance
Internal trade performance calculations