Estimation of Local Average Treatment Effect by Data Combination
Authors: Kazuhiko Shinoda, Takahiro Hoshino8295-8303
AAAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We test the performance of the proposed method with synthetic and real-world datasets. The details of the datasets and other setups can be found in Appendix D in the supplementary material. We denote the separate estimation as SEP. Synthetic Experiment We generated synthetic data with the different shapes of ยต, the covariates dimension qx, and the size of training samples. The mean and standard deviation of the MSE over 100 trials are summarized in Table 1. |
| Researcher Affiliation | Academia | 1Graduate School of Economics, Keio University 2Faculty of Economics, Keio University 3RIKEN AIP |
| Pseudocode | No | The paper describes mathematical formulations and steps but does not include any clearly labeled 'Pseudocode' or 'Algorithm' blocks. |
| Open Source Code | Yes | The proof can be found in Appendix B in the supplementary material. 1https://github.com/kazushino/AAAI22 |
| Open Datasets | Yes | We used the dataset of the National Job Training Partnership Act (JTPA) study. This is one of the largest RCT dataset for job training evaluations in the US with approximately 20000 participants, and it has been used in the several previous studies on causal inference (Bloom et al. 1997; Abadie, Angrist, and Imbens 2002; Donald, Hsu, and Lieli 2014). |
| Dataset Splits | Yes | For the model selection, we can choose a model that minimizes b J( bf, bg) evaluated on a validation set. |
| Hardware Specification | No | The paper does not provide specific details about the hardware used for its experiments. |
| Software Dependencies | No | The paper mentions software like 'Optuna: A Next-Generation Hyperparameter Optimization Framework' (Akiba et al. 2019) but does not provide specific version numbers for any software dependencies used in its experiments. |
| Experiment Setup | No | The details of the datasets and other setups can be found in Appendix D in the supplementary material. |