Metraipolis › archive › dc_36257eb5ec9a

Conditional independence testing through marginal permutation — arXiv:2611.03255

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.03255
License (as archived)
CC0-1.0
Archived
2026-11-03T18:43:44.878Z
SHA-256 of the archived text
1616e9edcf21ffcba1c61f19a527205dea8533d08a58ad6fb50ea0ecdfa5564d
Stable URL of this page
https://metraipolis.com/documento/dc_36257eb5ec9a

How to cite

Conditional independence testing through marginal permutation — arXiv:2611.03255 — arXiv (en), https://arxiv.org/abs/2611.03255 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_36257eb5ec9a (Metraipolis archive dc_36257eb5ec9a, sha256 1616e9edcf21ffcba1c61f19a527205dea8533d08a58ad6fb50ea0ecdfa5564d)
BibTeX
@misc{dc36257eb5ec9a,
  title = {Conditional independence testing through marginal permutation — arXiv:2611.03255},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.03255},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_36257eb5ec9a; sha256 1616e9edcf21ffcba1c61f19a527205dea8533d08a58ad6fb50ea0ecdfa5564d},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.03255. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Marginal permutation is exact for testing marginal independence but generally invalid for conditional independence because it breaks the dependence between the exposure ($\bm X$) and covariates ($\bm Z$). In this paper, we show that empirical studentization can nevertheless make this deliberately misspecified resampling scheme asymptotically valid for important conditional independence tests based on scores. The resulting procedures are novel in the sense that they are valid in finite samples when the exposure is marginally independent of the outcome ($\bm Y$) and covariates, and asymptotically valid under the broader conditional independence null, $\bm X\indep \bm Y\mid \bm Z$, subject to method-specific conditions. For generalized linear model scores, separately studentizing the observed and permuted scores yields validity under a correctly specified null mean despite \emph{arbitrary} conditional variance misspecification. Moreover, the permutation calibration also applies to tests based on generalized and projected covariance measures. Their validity accommodates machine learning nuisance estimates under a condition on the stability of refitting after permutation, for which we provide sufficient conditions for common regression learners. Permutation calibration preserves first-order local power while better capturing the finite-sample discreteness and skewness induced by sparse observations. Simulations and single-cell CRISPR screen analysis show substantial calibration gains with our methods when effective sample size is small.

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv stat.ME, nuovi lavori del giorno).
Categorie arXiv: Methodology (stat.ME) ; Statistics Theory (math.ST).
Pagina del lavoro: https://arxiv.org/abs/2611.03255 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-03T18:13:53.136Z dalla pagina https://arxiv.org/list/stat.ME/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).