generate.data3d {hero}

Generate 3d data

Description

Generate data related to Section 7.2 of Lu et al. (2012) (code from author). Define n = c(128, 128, 24). Then x[[i]] = (1:n[i])/n[i] - 1/2/n[i]. These are the observed data locations. For i, j, k spanning the full length of each element of x, mu3d[i, j, k] = x[[1]][i]^2 + x[[2]][j]^2 + x[[3]][k]^2. Lastly, data3d = mu3d + 0.5 * rnorm(n[1] * n[2] * n[3]).

Usage

generate.data3d()

generate_data3d()

generateData3d()

GenerateData3d()

Value

A list with components x, mu3d, and data3d. x is a list of sequences with length 128, 128, and 24. mu3d and data3d are arrays of size 128 by 128 by 24.

Author(s)

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

References

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

dat = generate.data3d()