Skip to content
DraconiticResearch
Search Sources API
203,234 studies31 sourcesUpdated 57 min ago
PreprintarXiv licencearXiv

Selective Randomization Inference for Interference

  • Takahide Yanagi
Open at sourcePDFDOICite

Abstract

Empirical researchers often use tests for interference as pretests or robustness checks. However, when the results guide the choice of models, parameters, or hypotheses, the resulting data-dependent selection can compromise the validity of subsequent inference. This paper develops a two-stage selective randomization inference framework for interference that explicitly conditions on the selection event to account for data-dependent selection. We first test the null hypothesis of no treatment spillover using a conditional randomization method. If this null is rejected, we test the null hypothesis of zero exposure contrast under a pre-specified exposure mapping that allows for spillover effects; otherwise, we test the null hypothesis of no direct effects under the assumption of no spillover. We prove that each second-stage test controls the selective Type I error rate in finite samples and evaluate the size and power of the proposed tests through Monte Carlo experiments.

Related

  1. Randomization Inference: Theory and Applications
    David M. Ritzwoller, Joseph P. Romano, Azeem M. Shaikh
    Journal of Political Economy MicroeconomicsPeer-reviewedLicence not statedFull text19 citedStatisticsOpenAlex
  2. Spatial causal inference in the presence of unmeasured confounding and interference
    Georgia Papadogeorgou, Srijata Samanta
    The Annals of Applied StatisticsPeer-reviewedLicence not statedFull text35 citedStatisticsOpenAlex
  3. Randomization Inference with Concentration Inequalities
    Tobias Freidling
    arXivPreprintCC BY 4.0Full textStatisticsarXiv
  4. Randomization Inference for Matched Pairs with Binary Outcomes
    Bob Wilson
    arXivPreprintarXiv licenceFull textStatisticsarXiv
  5. Selective Interference Suppression of Siamese-Net in Heterogeneous Interference Channels
    Arkadeep Sinha, Shubham Paul, R. Manivasakan +2
    arXivPreprintarXiv licenceFull textGeneral mathematicsarXiv

Cite

BibTeXRIS
BibTeX
@misc{yanagi2026selective,
  title = {{Selective Randomization Inference for Interference}},
  author = {Takahide Yanagi},
  howpublished = {arXiv},
  year = {2026},
  month = nov,
  doi = {10.48550/arxiv.2611.02443},
  eprint = {2611.02443},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2611.02443},
}
RIS
TY  - UNPB
TI  - Selective Randomization Inference for Interference
AU  - Takahide Yanagi
PY  - 2026
DA  - 2026/11/02
PB  - arXiv
DO  - 10.48550/arxiv.2611.02443
UR  - https://arxiv.org/abs/2611.02443
AB  - Empirical researchers often use tests for interference as pretests or robustness checks. However, when the results guide the choice of models, parameters, or hypotheses, the resulting data-dependent selection can compromise the validity of subsequent inference. This paper develops a two-stage selective randomization inference framework for interference that explicitly conditions on the selection event to account for data-dependent selection. We first test the null hypothesis of no treatment spillover using a conditional randomization method. If this null is rejected, we test the null hypothesis of zero exposure contrast under a pre-specified exposure mapping that allows for spillover effects; otherwise, we test the null hypothesis of no direct effects under the assumption of no spillover. We prove that each second-stage test controls the selective Type I error rate in finite samples and evaluate the size and power of the proposed tests through Monte Carlo experiments.
ER  -