| Title |
Interpretable machine learning for heterogeneous treatment effect estimators with Double ML: a case of access to credit for SMEs / |
| Authors |
Medianovskyi, Kyrylo ; Malakauskas, Aidas ; Lakstutiene, Ausrine ; Yahia, Sadok Ben |
| DOI |
10.1016/j.procs.2023.10.207 |
| Full Text |
|
| Is Part of |
Procedia computer science: 27th international conference on knowledge based and intelligent information and engineering sytems (KES 2023) / edited by Robert Howlett.. Amsterdam : Elsevier. 2023, vol. 225, p. 2163-2172.. ISSN 1877-0509 |
| Keywords [eng] |
interpretable machine learning ; explainable artificial intelligence ; double ML ; CATE ; SHAP ; partial dependence plot |
| Abstract [eng] |
Asymptotically consistent estimators of a treatment effect under many potential confounders became possible with the latest advancements in doubly-robust causal inference models (e.g., Double ML). In this study, we propose SAFE-TH framework to estimate and explain the heterogeneous treatment effect with partial dependence plots and report it under a reduced hypothesis space of interest. We analyze a shift in accessibility to credit for small to medium enterprises (SMEs) during the first months of the COVID-19 pandemic. Utilizing the proposed framework can improve the interpretability of CATE models by identifying and providing confidence intervals for regions of heterogeneity. |
| Published |
Amsterdam : Elsevier |
| Type |
Conference paper |
| Language |
English |
| Publication date |
2023 |
| CC license |
|
