ChatGPT Micro-Cap Experiment

Welcome to the repo behind my 6-month live trading experiment where ChatGPT manages a real-money micro-cap portfolio.

Overview on getting started: Here

Repository Structure

• trading_script.py - Main trading engine with portfolio management and stop-loss automation
• Scripts and CSV Files/ - My personal portfolio (updates every trading day)
• Start Your Own/ - Template files and guide for starting your own experiment
• Weekly Deep Research (MD|PDF)/ - Research summaries and performance reports
• Experiment Details/ - Documentation, methodology, prompts, and Q&A

The Concept

Every day, I kept seeing the same ad about having some A.I. pick undervalued stocks. It was obvious it was trying to get me to subscribe to some garbage, so I just rolled my eyes.
Then I started wondering, "How well would that actually work?"

So, starting with just $100, I wanted to answer a simple but powerful question:

Can powerful large language models like ChatGPT actually generate alpha (or at least make smart trading decisions) using real-time data?

Each trading day:

• I provide it trading data on the stocks in its portfolio.
• Strict stop-loss rules apply.
• Every week I allow it to use deep research to reevaluate its account.
• I track and publish performance data weekly on my blog: Here

Research & Documentation

• Research Index
• Disclaimer
• Q&A
• Prompts
• Starting Your Own
• Research Summaries (MD)
• Full Deep Research Reports (PDF)

Current Performance

Last Updated: August 2025

Current Status: Portfolio is outperforming the S&P 500 benchmark

Performance data is updated after each trading day. See the CSV files in Scripts and CSV Files/ for detailed daily tracking.

Features of This Repo

• Live trading scripts — used to evaluate prices and update holdings daily
• LLM-powered decision engine — ChatGPT picks the trades
• Performance tracking — CSVs with daily PnL, total equity, and trade history
• Visualization tools — Matplotlib graphs comparing ChatGPT vs. Index
• Logs & trade data — auto-saved logs for transparency

Why This Matters

AI is being hyped across every industry, but can it really manage money without guidance?

This project is an attempt to find out — with transparency, data, and a real budget.

Tech Stack & Features

Core Technologies

• Python - Core scripting and automation
• pandas + yFinance - Market data fetching and analysis
• Matplotlib - Performance visualization and charting
• ChatGPT-4 - AI-powered trading decision engine

Key Features

• Robust Data Sources - Yahoo Finance primary, Stooq fallback for reliability
• Automated Stop-Loss - Automatic position management with configurable stop-losses
• Interactive Trading - Market-on-Open (MOO) and limit order support
• Backtesting Support - ASOF_DATE override for historical analysis
• Performance Analytics - CAPM analysis, Sharpe/Sortino ratios, drawdown metrics
• Trade Logging - Complete transparency with detailed execution logs

System Requirements

• Python 3.7+
• Internet connection for market data
• ~10MB storage for CSV data files

Follow Along

The experiment runs from June 2025 to December 2025.
Every trading day I will update the portfolio CSV file.
If you feel inspired to do something similar, feel free to use this as a blueprint.

Updates are posted weekly on my blog — more coming soon!

One final shameless plug: (https://substack.com/@nathanbsmith?utm_source=edit-profile-page)

Find a mistake in the logs or have advice?
Please reach out here: [email protected]