LLMs for Time Series: an Application for Single Stocks and Statistical Arbitrage

TL;DR

This paper demonstrates that Large Language Models can predict nearly random financial time series and identify market inefficiencies, challenging the belief that they are unsuitable for such tasks.

Contribution

It provides the first evidence that LLMs can handle noisy financial data and generate alpha, showing potential beyond pattern recognition in structured data.

Findings

LLMs can predict near-noise time series effectively.

LLMs can identify market inefficiencies and generate alpha.

Significant room for improving LLM performance in finance.

Abstract

Recently, LLMs (Large Language Models) have been adapted for time series prediction with significant success in pattern recognition. However, the common belief is that these models are not suitable for predicting financial market returns, which are known to be almost random. We aim to challenge this misconception through a counterexample. Specifically, we utilized the Chronos model from Ansari et al.(2024) and tested both pretrained configurations and fine-tuned supervised forecasts on the largest American single stocks using data from Guijarro-Ordonnez et al.(2022). We constructed a long/short portfolio, and the performance simulation indicates that LLMs can in reality handle time series that are nearly indistinguishable from noise, demonstrating an ability to identify inefficiencies amidst randomness and generate alpha. Finally, we compared these results with those of specialized models and smaller deep learning models, highlighting significant room for improvement in LLM performance to further enhance their predictive capabilities.