generate.data3d {hero} | R Documentation |
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])
.
generate.data3d()
generate_data3d()
generateData3d()
GenerateData3d()
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.
Joshua French. Based off code by Luo Xiao (see 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>
dat = generate.data3d()