Data-driven trading using machine learning, deep learning, and AI-enhanced models.
ChatGPT can extract forward-looking investment signals from earnings calls. The paper shows that its investment score predicts firm capex up to 9 quarters ahead and negatively forecasts future stock returns—revealing information not yet priced by markets.
ChatGPT can classify crypto tweets into bullish or bearish and predict Bitcoin returns better than BERT or VADER. Its sentiment signals show strong predictive power for daily returns.
ChatGPT can predict future UK interest rate decisions by analyzing central bank speeches—turning policy tone into a reliable forecasting signal.
ChatGPT can turn bloated disclosures into sharp summaries. The sentiment from these summaries predicts stock returns better than traditional methods — revealing alpha in plain sight.
Can ChatGPT predict stocks? Yes—GPT-4 applied to headlines yields a high-Sharpe intraday strategy with 650%+ returns. Predictive power emerges only in large models.
This paper shows that ChatGPT-3.5 can predict aggregate U.S. stock returns using WSJ headlines. DeepSeek and smaller LLMs (like BERT) do not match this predictive performance.
This paper introduces a ChatGPT-based Commodity News Ratio Index (CNRI) to forecast excess returns on commodity futures. Using over 2.5 million articles across 18 commodities and 80 years, the CNRI significantly outperforms traditional predictors both in- and out-of-sample.
This paper compares machine learning methods to predict stock returns and measure equity risk premia. The best-performing model (a 3-layer neural net) earns out-of-sample Sharpe ratios as high as 2.45.
This paper introduces a new class of asset pricing models using autoencoders. By embedding firm characteristics into a neural network architecture, the authors construct a nonlinear conditional factor model that enforces no-arbitrage. It dramatically outperforms linear benchmarks.
Machine learning (ML) signals look great in backtests—but they often break when exposed to economic restrictions. This paper shows that ML-based strategies lose 50–80% of their alpha when excluding microcaps, distressed stocks, or accounting for transaction costs.
This paper develops the Policy Change Index (PCI) for China, a machine learning-based, language-free signal that predicts upcoming policy shifts by analyzing the page placement of articles in the People’s Daily. The PCI successfully anticipates major Chinese policy events.
This paper uses deep learning to estimate the stochastic discount factor (SDF) that prices all U.S. stocks. The authors combine a feedforward network, LSTM, and a generative adversarial network (GAN) to build a non-linear, no-arbitrage-compliant model.