Leave-one-out testing for node-level differences in Gaussian graphical models
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911392231063552 |
|---|---|
| author | Benussi, Davide Alongi, Ester Banzato, Erika |
| author_facet | Benussi, Davide Alongi, Ester Banzato, Erika |
| contents | We study two-sample equality testing in Gaussian graphical models. Classical likelihood ratio tests on decomposable graphs admit clique-wise factorizations, offering limited localization and unstable finite-sample behaviour. We propose node-level inference via a leave-one-out Bartlett-adjusted test on a fully connected graph. The resulting increments have standard chi-square null limits, enabling calibrated significance for single nodes and fixed-size subsets. Simulations confirm validity, and a case study shows practical utility. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_15896 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Leave-one-out testing for node-level differences in Gaussian graphical models Benussi, Davide Alongi, Ester Banzato, Erika Methodology We study two-sample equality testing in Gaussian graphical models. Classical likelihood ratio tests on decomposable graphs admit clique-wise factorizations, offering limited localization and unstable finite-sample behaviour. We propose node-level inference via a leave-one-out Bartlett-adjusted test on a fully connected graph. The resulting increments have standard chi-square null limits, enabling calibrated significance for single nodes and fixed-size subsets. Simulations confirm validity, and a case study shows practical utility. |
| title | Leave-one-out testing for node-level differences in Gaussian graphical models |
| topic | Methodology |
| url | https://arxiv.org/abs/2601.15896 |