dip {modeest}R Documentation

Compute Hartigan's DIP Test Statistic for Unimodality

Description

Computes Hartigan's DIP test statistic for testing unimodality, and additionally the modal interval. This function is taken from Martin Maechler's diptest package.

Usage

dip(x, 
    full.result = FALSE, 
    debug = FALSE)

Arguments

x numeric. Vector of observations.
full.result logical. If TRUE returns the full result list, see below.
debug logical. If TRUE, some tracing information is printed (from the C routine).

Value

Depending on full.result either a number, the dip statistic, or a list with components

x the sorted unname()d data.
n length(x).
dip the dip statistic
lo.hi indices into x for lower and higher end of modal interval
xl, xu lower and upper end of modal interval
gcm, lcm (last used) indices for greatest convex minorant and the least concave majorant.
mn, mj index vectors of length n for the GC minorant and the LC majorant respectively.

Note

For n <= 3 where n <- length(x), the dip statistic is always zero, i.e., there's no possible dip test.

Yong Lu lyongu+@cs.cmu.edu found in Oct 2003 that the code wasn't giving symmetric results for mirrored data (and was giving results of almost 1, and then found the reason, a misplaced ")" in the original Fortran code. This bug has been corrected for diptest version 0.25-0.

Author(s)

Martin Maechler maechler@stat.math.ethz.ch, based on earlier code from Dario Ringach dario@wotan.cns.nyu.edu

References

See Also

isoreg for isotonic regression.

Examples

data(precip)
plot(density(precip))
str(dip(precip, full = TRUE, debug = TRUE))

[Package modeest version 1.06 Index]