Séminaire de Probabilités et Statistique :

Le 29 mars 2021 à 13:45 - UM - Bât 09 - Salle de conférence (1er étage)

Présentée par Neuvial Pierre - Institut de Mathématiques de Toulouse

Post hoc inference for genomics and neuroimaging

We present confidence bounds on the false positives contained in subsets of selected null hypotheses. As the coverage probability holds simultaneously over all possible subsets, these bounds can be applied to an arbitrary number of possibly data-driven subsets.

We show how to construct and efficiently implement bounds that are adaptive to unknown dependence. We illustrate their application to two multiple two-sample settings: differential gene expression studies in genomics, and fMRI studies in neuroimaging. We also discuss connections with post hoc bounds based on closed testing.

Relevant papers:
- Blanchard G, Neuvial P, Roquain E: On agnostic post hoc approaches to false positive control. Book chapter to appear in Handbook of Multiple Comparisons. https://hal.archives-ouvertes.fr/hal-02320543
- Blanchard G, Neuvial P, Roquain E: Post Hoc Confidence Bounds on False Positives Using Reference Families. Annals of Statistics 48: 1281-1303, 2020 https://hal.archives-ouvertes.fr/hal-01483585

WEBINAIRE ouvert à toutes et tous : https://umontpellier-fr.zoom.us/j/85813807839