rh: Rotated H-transform

In hero: Spatio-Temporal (Hero) Sandwich Smoother

Rotated H-transform

Description

A rotation of the H-transform of the array a by a matrix x.

Usage

rh(x, a, transpose = FALSE)

Arguments

x	A matrix-like object. See Details.
a	An d-dimensional array
transpose	A logical value. The Default is FALSE. If TRUE, then the transpose of A

Details

x should be matrix-like. This function has been tested when x is a matrix object or a Matrix.

Assuming a is of size c_1 \times c_2 \times \dots \times c_d, then x is of size r \times c_1.

Value

A rotated, h-transformed array

Author(s)

Joshua French. Based off code by Luo Xiao (see References).

References

Currie, I. D., Durban, M. and Eilers, P. H. (2006), Generalized linear array models with applications to multidimensional smoothing. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 68: 259-280. <doi:10.1111/j.1467-9868.2006.00543.x>

Xiao, L. , Li, Y. and Ruppert, D. (2013), Fast bivariate P-splines: the sandwich smoother. J. R. Stat. Soc. B, 75: 577-599. <doi:10.1111/rssb.12007>

Examples

dim = c(10:12)
# construct random array
a = array(rnorm(prod(dim)), dim = dim)
# construct random matrix
x = matrix(rnorm(15 * dim[1]), nrow = 15)
rhxa = rh(x, a)

hero documentation built on July 26, 2023, 5:11 p.m.