[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-72008":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":15,"stars7d":17,"stars30d":18,"stars90d":16,"forks30d":16,"starsTrendScore":19,"compositeScore":20,"rankGlobal":10,"rankLanguage":10,"license":10,"archived":21,"fork":21,"defaultBranch":22,"hasWiki":23,"hasPages":21,"topics":24,"createdAt":10,"pushedAt":10,"updatedAt":25,"readmeContent":26,"aiSummary":27,"trendingCount":16,"starSnapshotCount":16,"syncStatus":28,"lastSyncTime":29,"discoverSource":30},72008,"LLM-Trading-Lab","LuckyOne7777\u002FLLM-Trading-Lab","LuckyOne7777","This repo powers my experiment where ChatGPT manages a real-money micro-cap stock portfolio.","",null,"Python",7462,1566,191,1,0,4,7,3,40.59,false,"main",true,[],"2026-06-12 02:02:57","# LLM Trading Lab\nThis repository started as a **6-month live micro-cap trading experiment** in which a large language model (ChatGPT) manages a real-money portfolio under strict, predefined rules.\n\nWhat began as a single experiment has evolved into a **baseline framework** for studying how large language models behave as portfolio decision-makers.  \nAll historical data, research artifacts, and logs are preserved for transparency and auditability.\n\n**Full research evaluation out now: [Evaluating ChatGPT as a Portfolio Decision-Maker in Micro-Cap Equities](Experiments\u002Fchatgpt_micro-cap\u002Fevaluation\u002Fpaper.pdf)**\n\n---\n\n## Running Your Own Experiment\n\nIf you want to run your own AI-managed trading experiment, check out this framework I created for LLM research: [LLM Investor Behavior Benchmark - LIBB](https:\u002F\u002Fgithub.com\u002FLuckyOne7777\u002FLLM-Investor-Behavior-Benchmark)\n\n## Repository Purpose\n\nThis repository serves two primary purposes:\n\n1. A **complete, forward-only record** of a live AI-managed trading experiment  \n2. A **reusable foundation** for future AI-driven trading experiments built on the same structure\n\nHistorical artifacts remain unchanged. New experiments, analyses, and methodologies are layered on top without rewriting past results.\n\n---\n\n```text\nChatGPT-Micro-Cap-Experiment\u002F\n│\n├─ README.md\n├─ requirements.txt\n├─ Makefile\n│\n├─ experiments\u002F\n│  └─ chatgpt_micro_cap\u002F\n│     │\n│     ├─ trading_script.py\n│     │\n|     ├─ graphing\u002F\n|     │  ├─ daily_returns.py\n|     │  ├─ drawdown.py\n|     │  └─ ...\n|     │ \n│     ├─ csv_files\u002F\n│     │  ├─ Daily_Updates.csv\n│     │  └─ Trade_Log.csv\n│     │\n│     ├─ evaluation\u002F\n│     │  ├─ evaluation_report.md\n│     │  └─ paper.pdf\n│     │\n│     ├─ collected_artifacts\u002F\n│     │  ├─ deep_research_index.md\n│     │  ├─ chats.md\n│     │  │\n│     │  ├─ Weekly_Deep_Research_MD\u002F\n│     │  │  ├─ Week_01_Summary.md\n│     │  │  ├─ Week_02_Summary.md\n│     │  │  └─ ...\n│     │  │\n│     │  └─ Weekly_Deep_Research_PDF\u002F\n│     │     ├─ Starting_Research.pdf\n│     │     ├─ Week_01.pdf\n│     │     ├─ Week_02.pdf\n│     │     └─ ...\n│     │\n│     ├─ images\u002F\n│     │  ├─ equity_vs_baseline.png\n│     │  ├─ repeated_exposure.png\n│     │  └─ ...\n│     │\n│     ├─ tables\u002F\n│     │  └─ metrics.txt\n│     │\n│     ├─ metrics\u002F\n│     │  ├─ load_dataV3.py\n│     │  └─ episode_pcr.py\n│     │\n│     └─ processing\u002F\n│        ├─ ProcessPortfolio.py\n|\n│\n├─\n\n```\n---\n\n## The Concept\n\nEvery 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. \nThen I started wondering, \"How well would that actually work?\" \n\nSo, 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?**\n\nToday, this repo has evolved into so much more than simply chasing alpha.\n\n---\n\n## Why This Matters\n\nAI is being aggressively marketed as a replacement for human decision-making across industries.  \nTrading is a domain where mistakes are measurable, irreversible, and costly.\n\nThis platform tests those claims using:\n\n- Forward-only decisions\n- Full transparency  \n- Publicly logged results\n\n---\n\n## Research & Documentation\n\nHere are the artifacts links for the Micro-Cap Experiment:\n\n- **Research Index:** [Deep Research Index](Experiments\u002Fchatgpt_micro-cap\u002Fcollected_artifacts\u002Fdeep_research_index.md)\n\n- **Decision Logs \u002F Chats:** [Chats](Experiments\u002Fchatgpt_micro-cap\u002Fcollected_artifacts\u002Fchats.md)\n\n---\n\n## Features of This Repository\n\n- 40 page PDF evaluation over results\n- Live trading engine used in production  \n- LLM-driven trade selection under hard constraints  \n- Daily CSV-based portfolio accounting  \n- Automated stop-loss enforcement  \n- Benchmark comparisons (S&P 500, Russell 2000)  \n- CAPM, Sharpe, Sortino, and drawdown analytics  \n- Full trade and decision logs\n\n---\n\n## Tech Stack\n\n- Python 3.11+  \n- pandas  \n- yfinance (primary data source)  \n- Stooq (fallback data source)  \n- Matplotlib  \n\n---\n\n## Future Work\n\nI am currently designing the future experiment over newly listed IPOs with monthly analysis on my [Substack](https:\u002F\u002Fnathanbsmith729.substack.com\u002F).\n\nAlso, I developing the general experimental framework I created for LLM research [LIBB](https:\u002F\u002Fgithub.com\u002FLuckyOne7777\u002FLLM-Investor-Behavior-Benchmark) for the upcoming and all future experiments.\n\n---\n\n## Contributing\n\nContributions are welcome.\n\n- Issues: bugs, edge cases, or design critiques  \n- Pull Requests: improvements, refactors, or extensions  \n- Collaboration: high-quality contributors may be invited to help maintain future experiments\n\nContributing guide:  \nhttps:\u002F\u002Fgithub.com\u002FLuckyOne7777\u002FChatGPT-Micro-Cap-Experiment\u002Fblob\u002Fmain\u002FOther\u002FCONTRIBUTING.md\n\n---\n\n## Contact\n\nAll my links can be found on my profile, feel free to reach out anywhere!\n","该项目通过让ChatGPT管理一个真实的微盘股票投资组合，进行了一项为期6个月的实验。核心功能包括使用大型语言模型作为投资决策者，并记录所有交易活动和研究结果以确保透明度和可审计性。技术上，项目基于Python构建，提供了完整的交易脚本、数据分析工具以及详尽的日志文件。适用于希望探索AI在金融领域应用的研究人员或投资者，特别是那些对利用人工智能进行股票交易感兴趣的人士。此外，它还为后续类似实验提供了一个可复用的基础框架。",2,"2026-06-11 03:39:56","high_star"]