Paper: arXiv 2404.05101

Abstract

This paper introduces StockGPT, an autoregressive ``number’’ model trained and tested on 70 million daily U.S.\ stock returns over nearly 100 years. Treating each return series as a sequence of tokens, StockGPT automatically learns the hidden patterns predictive of future returns via its attention mechanism. On a held-out test sample from 2001 to 2023, daily and monthly rebalanced long-short portfolios formed from StockGPT predictions yield strong performance. The StockGPT-based portfolios span momentum and long-/short-term reversals, eliminating the need for manually crafted price-based strategies, and yield highly significant alphas against leading stock market factors, suggesting a novel AI pricing effect. This highlights the immense promise of generative AI in surpassing human in making complex financial investment decisions.

Complexity vs Empirical Score

  • Math Complexity: 7.5/10
  • Empirical Rigor: 6.0/10 (capped from 9.0: transaction costs only partially modelled; see penalty flags)
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper employs advanced generative AI/transformer architectures with attention mechanisms (high math complexity) and demonstrates rigorous out-of-sample backtesting across 23 years with 70M data points, detailed portfolio construction, and statistical significance testing (high empirical rigor).

Research Flowchart

  flowchart TD
  A["Research Goal: Can Generative AI<br>surpass human stock prediction?"] --> B["Methodology: Autoregressive<br>Attention-Based Model"]
  B --> C["Input: 70M Daily US Stock<br>Returns (1926-2023)"]
  C --> D["Process: Train StockGPT to<br>predict next return tokens"]
  D --> E["Output: Daily/Monthly<br>Long-Short Portfolios"]
  E --> F["Key Findings:<br>Strong Out-of-Sample Returns<br>Significant Alphas vs. Factors<br>Novel AI Pricing Effect"]

Cited References (31)

  1. The social impact of generative ai: An analysis on ChatGPT (2023)
  2. Generative AI: Here to stay, but for good? (2023)
  3. Artificial intelligence in developing countries: The impact of generative artificial intelligence (AI) technologies for development (2023)
  4. Sentiment Trading with Large Language Models — → 2412.19245
  5. Expected returns and large language models (2022)
  6. Can ChatGPT forecast stock price movements? Return predictability and large language models — → 2304.07619
  7. Attention is all you need (2017)
  8. Language models are few-shot learners (2020)
and 23 more
  1. BERT: Pre-training of deep bidirectional transformers for language understanding (2018) — link
  2. OPT: Open pre-trained transformer language models (2022) — link
  3. BloombergGPT: A large language model for finance (2023) — link
  4. FinBERT: A Pretrained Language Model for Financial Communications (2020) — link
  5. Financial machine learning (2023)
  6. ClinicalGPT: large language models finetuned with diverse medical data and comprehensive evaluation (2023) — link
  7. Chronos: Learning the Language of Time Series (2024) — link
  8. Empirical asset pricing via machine learning (2020)
  9. (Re-) Imag (in) ing Price Trends (2023)
  10. Risk, return, and equilibrium: Empirical tests (1973)
  11. Hypothesis testing with efficient method of moments estimation (1987)
  12. A five-factor asset pricing model (2015)
  13. An augmented q-factor model with expected growth (2021)
  14. Does the stock market overreact? (1985)
  15. Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency (1993)
  16. Evidence of predictable behavior of security returns (1990)
  17. A new anomaly: The cross-sectional profitability of technical analysis (2013)
  18. Efficient capital markets (1970)
  19. Experimental evidence on the productivity effects of generative artificial intelligence (2023)
  20. Navigating the jagged technological frontier: Field experimental evidence of the effects of AI on knowledge worker productivity and quality (2023)
  21. Deep order flow imbalance: Extracting alpha at multiple horizons from the limit order book (2023)
  22. Investor sentiment and the cross-section of stock returns (2006)
  23. The uneven impact of generative AI on entrepreneurial performance (2023)

2 of 31 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.