Generate two-dimensional data related to the f1 function
of Lu et al. (2012) (code from author). Define n =
c(60, 80). Then x[[i]] = (1:n[i])/n[i] -
1/2/n[i]. These are the observed data locations. For
i and j spanning the full length of each
element of x, mu2d[i, j] = sin(2 * pi *
(x[[1]][i] - .5) ^ 3) * cos(4 * pi * x[[2]][j]). Lastly,
data2d = mu2d + rnorm(prod(n)).
Usage
generate.data2d()
generate_data2d()
generateData2d()
GenerateData2d()
Arguments
Value
A list with components x, mu2d, and
data2d. x is a list of sequences with
length 60 and 80. mu2d and data2d are
matrices of size 60 by 80.
Author
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>