English

Multiple testing via $FDR_L$ for large-scale imaging data

Statistics Theory 2011-03-11 v1 Statistics Theory

Abstract

The multiple testing procedure plays an important role in detecting the presence of spatial signals for large-scale imaging data. Typically, the spatial signals are sparse but clustered. This paper provides empirical evidence that for a range of commonly used control levels, the conventional FDR\operatorname {FDR} procedure can lack the ability to detect statistical significance, even if the pp-values under the true null hypotheses are independent and uniformly distributed; more generally, ignoring the neighboring information of spatially structured data will tend to diminish the detection effectiveness of the FDR\operatorname {FDR} procedure. This paper first introduces a scalar quantity to characterize the extent to which the "lack of identification phenomenon" (LIP\operatorname {LIP}) of the FDR\operatorname {FDR} procedure occurs. Second, we propose a new multiple comparison procedure, called FDRL\operatorname {FDR}_L, to accommodate the spatial information of neighboring pp-values, via a local aggregation of pp-values. Theoretical properties of the FDRL\operatorname {FDR}_L procedure are investigated under weak dependence of pp-values. It is shown that the FDRL\operatorname {FDR}_L procedure alleviates the LIP\operatorname {LIP} of the FDR\operatorname {FDR} procedure, thus substantially facilitating the selection of more stringent control levels. Simulation evaluations indicate that the FDRL\operatorname {FDR}_L procedure improves the detection sensitivity of the FDR\operatorname {FDR} procedure with little loss in detection specificity. The computational simplicity and detection effectiveness of the FDRL\operatorname {FDR}_L procedure are illustrated through a real brain fMRI dataset.

Keywords

Cite

@article{arxiv.1103.1966,
  title  = {Multiple testing via $FDR_L$ for large-scale imaging data},
  author = {Chunming Zhang and Jianqing Fan and Tao Yu},
  journal= {arXiv preprint arXiv:1103.1966},
  year   = {2011}
}

Comments

Published in at http://dx.doi.org/10.1214/10-AOS848 the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)

R2 v1 2026-06-21T17:37:42.900Z