This is the official repository for Empirical Asset Pricing with Large Language Model Agents.

1. First clone the directory.

git submodule init; git submodule update

(If showing error of no permission, need to first add a new SSH key to your GitHub account.)

2. Install dependencies.

Create a new environment using conda, with Python >= 3.10.6 Install PyTorch (version >= 2.0.0). The repo is tested with PyTorch version of 1.10.1 and there is no guarentee that other version works. Then install other dependencies via:

pip install -r requirements.txt

3. Download news dataset.

Download the WSJ dataset and unzip it.

1. Build the Factor dataset from https://github.com/bkelly-lab/ReplicationCrisis with your CRSP credential, and download the daily return data from https://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html for free.

2. Setup your configurations in config.yaml and input your API keys accordingly.

3. Use analysis.py to produce the analysis report features.

4. Use model.py to train the hybrid asset pricing model.

If you use this code in your research, please cite the following paper:

@inproceedings{cheng2025empiricalassetpricinglarge,
      title={Empirical Asset Pricing with Large Language Model Agents}, 
      author={Junyan Cheng and Peter Chin},
      year={2025},
      maintitle={The Thirteenth International Conference on Learning Representations (ICLR)},
      booktitle={Advances in Financial AI Workshop},
}