Directional Regularity for Multivariate Functional Data
direg is a R package for estimating the directional regularity of multivariate functional data. It allows one to estimate the change-of-basis matrix
to take advantage of the intrinsic anisotropy of a process. Additional features include simulating anisotropic fractional brownian motions (fBms), surface smoothing
and anisotropic detection. Currently implementated for bi-variate processes.
To install the latest version from Github, please use
devtools::install_github("sunnywang93/direg")
Please note that R version >=4.1.0 is required.
The methodological and theoretical aspects of the paper can be found at: