The DeLIVR Estimator is a Python-based tool that leverages deep learning for instrumental variable regression. This project aims to the performance of DeLiVR and provide an easy-to-use interface for academic researchers.
- Navigate to Stage1_DeLiVR.ipynb and Stage2_DeLiVR.ipynb on GitHub.
- Click on the Open in Colab button.
- Download the data files from the data folder.
- Upload these files to Google Drive for use in Colab notebooks.
- Open Stage1_DeLiVR.ipynb and execute all cells to simulate data, where "3" in filenames like s1_Z_3.csv denotes a cubic relationship.
- Save the generated data (s1_Z_3.csv, s2_Z_3.csv, etc.) to your Google Drive for analysis.
- Handles various types of relationships (linear, quadratic, cubic)
Contributions are welcome! Please fork the repository and submit a pull request with your changes.