Given the following R code:
require(lattice)
x <- c(1:10)
y <- c(1:10)
g <- expand.grid(x = 1:10, y = 1:10, gr = 1:2)
g$z <- c(as.vector(outer(x,y,"*")), rep(50,100))
wireframe(z ~ x * y, data = g, groups = gr)
The intersection of the resulting surfaces is ugly, since it follows the grid-lines.
Is there a way to make the intersection between the surfaces look nicer (besides increasing the resolution of the grid)? Maybe by passing some parameters or using another package for visualization?
Well, since I couldn't resist goofing off, here are a few possiblities for ways to smooth your data.
The package
scvm
appears to have some 2D model-fitting tools.The
fields
package is recommended here: How can I smooth an array in R?The
DiceKriging
package is reviewed here: https://stats.stackexchange.com/questions/13510/fitting-multivariate-natural-cubic-spline