Spatiotemporal Hypergraph Convolution Network for Stock Movement Forecasting

This codebase contains the python scripts for STHGCN, the model for the ICDM 2020 paper link.

Environment & Installation Steps

Python 3.6, Pytorch, Pytorch-Geometric and networkx.

Dataset and Preprocessing

Download the dataset and follow preprocessing steps from here.

bash download.sh

Run

Execute the following python command to train STHGCN:

make test_phase=1 save_dir=save

test_phase : phase that you want to test

Cite

Consider citing our work if you use our codebase

@INPROCEEDINGS{9338303,  author={Sawhney, Ramit and Agarwal, Shivam and Wadhwa, Arnav and Shah, Rajiv Ratn},  booktitle={2020 IEEE International Conference on Data Mining (ICDM)},   title={Spatiotemporal Hypergraph Convolution Network for Stock Movement Forecasting},   year={2020},  volume={},  number={},  pages={482-491},  doi={10.1109/ICDM50108.2020.00057}}