Package pbj details

Parametric Bootstrap Joint Testing Procedures for Neuroimaging

These parametric bootstrap joint (PBJ) inference procedures are designed for the analysis of neuroimaging data. The statistical tools are more generally applicable, but this package is designed to allow input and output data for (Neuroimaging Informatics Technology Initiative) NIfTI images. The PBJ tools are designed for voxel-wise and cluster-extent hypothesis testing methods and include semi-PBJ (sPBJ) inference that is robust to variance misspecification using an estimating equations approach. For details, see Vandekar, Simon N., Satterthwaite, Theodore D., Xia, Cedric H., Ruparel, Kosha, Gur, Ruben C., Gur, Raquel E., Shinohara, Russell T. 2019. "Robust Spatial Extent Inference with a Semiparametric Bootstrap Joint Testing Procedure". Biometrics. (in press) and Vandekar, Simon N., Satterthwaite, Theodore D. and Rosen, Adon and Ciric, Rastko and Roalf, David R. and Ruparel, Kosha and Gur, Ruben C. and Gur, Raquel E. and Shinohara, Russell T.. 2018. "Faster family-wise error control for neuroimaging with a parametric bootstrap". Biostatistics. 19(4):497-513.

Maintainer: Simon Vandekar < simon.vandekar at vanderbilt.edu >

Documentation

From within R, enter citation('pbj')

If you have any problems with this package you can open a new issue or check the already existing ones here.

To install this package, start R and enter:

source("https://neuroconductor.org/neurocLite.R")

# Default Install
neuro_install('pbj')

# from GitHub
neuro_install('pbj', release = "stable", release_repo = "github")
neuro_install('pbj', release = "current", release_repo = "github")

More detailed installation instructions can be found here.