Smoothing z-values according to their x, y positions
Dear David, Thanks a lot for pointing out kde2d, just tried it out but the problem is that it indeed takes the density of points into account, which I dont want. For example, if in an region of surface S I've got 10,000 points, and that their average height is 0.5, and in an other region I've got only ten points and their average value if also 0.5, I'd like all these points to be transformed to the same 0.5 value. At the moment, it seems that it's not the case. For example, the range of the values I give is: 0.2 - 3.7, but the range of the values that are outputed is 0 - 0.17. I haven't looked yet at the locfit package as it is not installed, but I will check it out! Thanks for helping! Emmanuel
On 20/03/2008, David Winsemius <dwinsemius at comcast.net> wrote:
"Emmanuel Levy" <emmanuel.levy at gmail.com> wrote in news:e4654710803191554k4dfc2728g70bde35f41627754 at mail.gmail.com:
> Dear Bert,
>
> Thanks for your reply - I indeed saw a lot of functions using:
> help.search("smooth")
>
> The problem is that most seem to not be very appropriate to what I'd
> like, or they seem extremely complicated (e.g. gma). I am probably
> missing something as I don't
> see how to use Loess. From my poor understanding, it seems to be for
> 2D data. Here I want to smooth the third "z" component.
>
> In the meantime I adapted a function from Greg Warnes, which sort of
> does the job (although the smoothing is not very nice).
> So I'd be very happy to learn how do do something similar with
> loess!
>
You could consider at combining the kde2d() function in the MASS package
within the VR bundle in combination with contour(). The example I
followed came from MASS 2ed in chapter 5's section on density estimation.
It is probably easiest to just copy the help example:
library(MASS)
attach(geyser)
f1 <- kde2d(duration[-272], duration[-1],
h = rep(1.5, 2), n = 50, lims = c(0.5, 6, 0.5, 6))
contour(f1, xlab = "previous duration",
ylab = "duration", levels = c(0.05, 0.1, 0.2, 0.4) )
I have also had success with Loader's locfit package.
--
David Winsemius
______________________________________________ R-help at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.