EUVIMEDAtrasti EUVIMED
Crossref

Addressing positivity violations in causal effect estimation using Gaussian process priors

Angela Yaqian Zhu iD, Nandita Mitra, Jason Roy

Statistics in Medicine · 2023-01-15 · 42 (1) · 33-51

DOI: 10.1002/sim.9600 ↗

At a glance

Journal article in Statistics in Medicine, published online on 2022-11-06. Written by Angela Yaqian Zhu and 2 co-authors. Institution: Janssen Research & Development, Johnson & Johnson, Raritan, New Jersey. Major MEDLINE topics: Models, Statistical. Cited 12 times so far (Crossref); 40 references.

SantraukaSubjectsBibliografiniai duomenysCiteRelated studies

Abstract (original language)

In observational studies, causal inference relies on several key identifying assumptions. One identifiability condition is the positivity assumption, which requires the probability of treatment be bounded away from 0 and 1. That is, for every covariate combination, it should be possible to observe both treated and control subjects the covariate distributions should overlap between treatment arms. If the positivity assumption is violated, population‐level causal inference necessarily involves some extrapolation. Ideally, a greater amount of uncertainty about the causal effect estimate should be reflected in such situations. With that goal in mind, we construct a Gaussian process model for estimating treatment effects in the presence of practical violations of positivity. Advantages of our method include minimal distributional assumptions, a cohesive model for estimating treatment effects, and more uncertainty associated with areas in the covariate space where there is less overlap. We assess the performance of our approach with respect to bias and efficiency using simulation studies. The method is then applied to a study of critically ill female patients to examine the effect of undergoing right heart catheterization.

Subjects

MeSH terms (MEDLINE)

Models, StatisticalHumansFemaleProbabilityComputer SimulationBias

Author keywords

Bayesian nonparametricsMetropolis within Gibbscausal inferenceextrapolationoverlappopulation average treatment effect

Authors and institutions

  • Janssen Research & Development, Johnson & Johnson, Raritan, New Jersey.
  • Department of Biostatistics, Epidemiology, and Informatics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania.
  • Department of Biostatistics and Epidemiology, School of Public Health, Rutgers University, Piscataway, New Jersey.

ORCID profiles: Angela Yaqian Zhu

Cite

APA

Zhu, A. Y., Mitra, N., & Roy, J. (2023). Addressing positivity violations in causal effect estimation using Gaussian process priors. Statistics in Medicine, 42(1), 33-51. https://doi.org/10.1002/sim.9600

Vancouver

Zhu AY, Mitra N, Roy J. Addressing positivity violations in causal effect estimation using Gaussian process priors. Statistics in Medicine. 2023;42(1):33-51. doi:10.1002/sim.9600

BibTeX
@article{Zhu2023,
  author = {Zhu, Angela Yaqian and Mitra, Nandita and Roy, Jason},
  title = {Addressing positivity violations in causal effect estimation using Gaussian process priors},
  journal = {Statistics in Medicine},
  year = {2023},
  volume = {42},
  number = {1},
  pages = {33--51},
  doi = {10.1002/sim.9600}
}
Start for free →