Displaying 1 result from an estimated 1 matches for "nbmodes".
Did you mean:
nb_modes
2003 Jul 17
0
Silverman modality test
...(1980), 42-56). Values for
the critical window width seem OK, which is not the case for the
significance levels.
If someone could give me a hint about what is wrong...
Or perhaps someone has already done a real implementation of this test?
Thanks,
Jerome Sackur
Inserm U562
Orsay, France
nbmodes <- function (x, h)
# returns how many modes there are in the kernel density estimate of
# vector x, with window width h.
{
modes <- density (x, bw=h)
modes <- diff (diff (modes$y) / abs (diff (modes$y)))
modes <- rep(1, length(modes))[modes==-2]
modes <- sum (modes)...