EUVIMEDExplore EUVIMED
Open access

Practical Considerations for Gaussian Process Modeling for Causal Inference in Quasi‐Experimental Studies With Panel Data

Sofia L Vega iD, Rachel C Nethery iD

Statistics in Medicine · 2026-07 · 45 (15-17)

DOI: 10.1002/sim.70652 ↗

At a glance

Journal article in Statistics in Medicine, published online on 2026-07-01. Written by Sofia L Vega and 1 co-authors. Institution: Department of Biostatistics, Harvard T.H. Chan School of Public Health, Boston, Massachusetts, USA. Major MEDLINE topics: Models, Statistical, Causality. 32 references. The full text is freely available in PubMed Central. Funded by: NIEHS NIH HHS, NIH HHS, Harvard Climate Change Solutions Fund.

AbstractSubjectsBibliographic detailsCiteRelated studies

Abstract (original language)

Estimating causal effects in quasi‐experiments with spatio‐temporal panel data often requires adjusting for unmeasured confounding that varies across space and time. Gaussian processes (GPs) offer a flexible, nonparametric modeling approach that can account for such complex dependencies through carefully chosen covariance kernels. In this paper, we provide a practical and interpretable framework for applying GPs to causal inference in panel data settings. We demonstrate how GPs generalize popular methods such as synthetic control and vertical regression, and we show that the GP posterior mean can be represented as a weighted average of observed outcomes, where the weights reflect spatial and temporal similarity. To support applied use, we explore how different kernel choices impact both estimation performance and interpretability, offering guidance for selecting between separable and nonseparable kernels. Through simulations and application to Hurricane Katrina mortality data, we illustrate how GP models can be used to estimate counterfactual outcomes and quantify treatment effects. All code and materials are made publicly available to support reproducibility and encourage adoption. Our results suggest that GPs are a promising and interpretable tool for addressing unmeasured spatio‐temporal confounding in quasi‐experimental studies.

Subjects

MeSH terms (MEDLINE)

Models, StatisticalCausalityHumansNormal DistributionComputer SimulationData Interpretation, Statistical

Author keywords

counterfactual predictionkernel selectionpolicy evaluationspatio‐temporal dataunmeasured confounding

Authors and institutions

  • Department of Biostatistics, Harvard T.H. Chan School of Public Health, Boston, Massachusetts, USA.

ORCID profiles: Sofia L Vega, Rachel C Nethery

Funding

  • NIEHS NIH HHS
  • NIH HHS
  • Harvard Climate Change Solutions Fund
  • Harvard Milton Fund
  • National Institutes of Health

Cite

APA

Vega, S. L., & Nethery, R. C. (2026). Practical Considerations for Gaussian Process Modeling for Causal Inference in Quasi‐Experimental Studies With Panel Data. Statistics in Medicine, 45(15-17). https://doi.org/10.1002/sim.70652

Vancouver

Vega SL, Nethery RC. Practical Considerations for Gaussian Process Modeling for Causal Inference in Quasi‐Experimental Studies With Panel Data. Statistics in Medicine. 2026;45(15-17). doi:10.1002/sim.70652

BibTeX
@article{Vega2026,
  author = {Vega, Sofia L and Nethery, Rachel C},
  title = {Practical Considerations for Gaussian Process Modeling for Causal Inference in Quasi‐Experimental Studies With Panel Data},
  journal = {Statistics in Medicine},
  year = {2026},
  volume = {45},
  number = {15-17},
  doi = {10.1002/sim.70652}
}
Start for free →