vows-package: Voxelwise semiparametrics
Description
This package efficiently performs inference on a large set of parametric or
semiparametric regressions that are "parallel" in the sense that they have a
common design matrix. The functions are inspired by neuroimaging
applications, where the parallel models pertain to a grid of brain locations
known as voxels.
Details
Functions ending in ".mp" ("massively parallel") are designed for responses
in the form of a (wide) matrix; functions ending in "4d" take
four-dimensional response data (e.g., a set of images) and convert it to
matrix form so that the corresponding ".mp" function can be applied.
Examples include lm.mp
and lm4d
for ordinary
linear models, rlrt.mp
and rlrt4d
for restricted
likelihood ratio tests (RLRTs) of a parametric null hypothesis vs. a smooth
alternative, and semipar.mp
and semipar4d
for
smoothing (see Reiss et al., 2014).
References
Reiss, P. T., Huang, L., Chen, Y.-H., Huo, L., Tarpey, T., and
Mennes, M. (2014). Massively parallel nonparametric regression, with an
application to developmental brain mapping. Journal of Computational
and Graphical Statistics, Journal of Computational and Graphical
Statistics, 23(1), 232--248.