Local Discovery by Partitioning: Polynomial-Time Causal Discovery Around Exposure-Outcome Pairs
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2023
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866913372324233216 |
|---|---|
| author | Maasch, Jacqueline Pan, Weishen Gupta, Shantanu Kuleshov, Volodymyr Gan, Kyra Wang, Fei |
| author_facet | Maasch, Jacqueline Pan, Weishen Gupta, Shantanu Kuleshov, Volodymyr Gan, Kyra Wang, Fei |
| contents | Causal discovery is crucial for causal inference in observational studies, as it can enable the identification of valid adjustment sets (VAS) for unbiased effect estimation. However, global causal discovery is notoriously hard in the nonparametric setting, with exponential time and sample complexity in the worst case. To address this, we propose local discovery by partitioning (LDP): a local causal discovery method that is tailored for downstream inference tasks without requiring parametric and pretreatment assumptions. LDP is a constraint-based procedure that returns a VAS for an exposure-outcome pair under latent confounding, given sufficient conditions. The total number of independence tests performed is worst-case quadratic with respect to the cardinality of the variable set. Asymptotic theoretical guarantees are numerically validated on synthetic graphs. Adjustment sets from LDP yield less biased and more precise average treatment effect estimates than baseline discovery algorithms, with LDP outperforming on confounder recall, runtime, and test count for VAS discovery. Notably, LDP ran at least 1300x faster than baselines on a benchmark. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2310_17816 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Local Discovery by Partitioning: Polynomial-Time Causal Discovery Around Exposure-Outcome Pairs Maasch, Jacqueline Pan, Weishen Gupta, Shantanu Kuleshov, Volodymyr Gan, Kyra Wang, Fei Machine Learning Methodology Causal discovery is crucial for causal inference in observational studies, as it can enable the identification of valid adjustment sets (VAS) for unbiased effect estimation. However, global causal discovery is notoriously hard in the nonparametric setting, with exponential time and sample complexity in the worst case. To address this, we propose local discovery by partitioning (LDP): a local causal discovery method that is tailored for downstream inference tasks without requiring parametric and pretreatment assumptions. LDP is a constraint-based procedure that returns a VAS for an exposure-outcome pair under latent confounding, given sufficient conditions. The total number of independence tests performed is worst-case quadratic with respect to the cardinality of the variable set. Asymptotic theoretical guarantees are numerically validated on synthetic graphs. Adjustment sets from LDP yield less biased and more precise average treatment effect estimates than baseline discovery algorithms, with LDP outperforming on confounder recall, runtime, and test count for VAS discovery. Notably, LDP ran at least 1300x faster than baselines on a benchmark. |
| title | Local Discovery by Partitioning: Polynomial-Time Causal Discovery Around Exposure-Outcome Pairs |
| topic | Machine Learning Methodology |
| url | https://arxiv.org/abs/2310.17816 |