EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

On the Classification–Causal Tradeoff in Neural Network Propensity Score Estimation

Seungman Kim, Jaehoon Lee, Kwanghee Jung

Stats · 31.03.2026 · 9 (2) · 37

DOI: 10.3390/stats9020037 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Stats, online erschienen am 31.03.2026. Verfasst von Seungman Kim und 2 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Observational studies serve as a vital alternative to randomized experiments but are highly susceptible to selection bias. Propensity score (PS) methods address this by balancing covariates between groups. Although including all relevant covariates is theoretically ideal, high dimensionality often destabilizes traditional estimation models. This study evaluates the efficacy of deep neural networks (DNN) and convolutional neural networks (CNN) for PS estimation compared to traditional logistic regression (LR), leveraging their capacity to handle complex nonlinear relationships and interactions. Using a Monte Carlo simulation across 36 conditions, model performance was evaluated based on bias and imbalance reduction. Results indicate that DNNs and CNNs significantly outperform LR. Specifically, while LR increased outcome bias by 17% and reduced covariate imbalance by only 5%, DNNs and CNNs reduced outcome bias by 13% and 16%, respectively, while decreasing covariate imbalance by 18% and 21%. We conclude that despite requiring specialized computational resources, neural networks offer substantial advantages for high-dimensional PS estimation. However, their reliable application necessitates stability-aware training and proper error rate thresholds to prevent probability degeneracy.

Zitieren

APA

Kim, S., Lee, J., & Jung, K. (2026). On the Classification–Causal Tradeoff in Neural Network Propensity Score Estimation. Stats, 9(2), 37. https://doi.org/10.3390/stats9020037

Vancouver

Kim S, Lee J, Jung K. On the Classification–Causal Tradeoff in Neural Network Propensity Score Estimation. Stats. 2026;9(2):37. doi:10.3390/stats9020037

BibTeX
@article{Kim2026,
  author = {Kim, Seungman and Lee, Jaehoon and Jung, Kwanghee},
  title = {On the Classification–Causal Tradeoff in Neural Network Propensity Score Estimation},
  journal = {Stats},
  year = {2026},
  volume = {9},
  number = {2},
  pages = {37},
  doi = {10.3390/stats9020037}
}
Kostenlos starten →