Simulates p-value for dip test (see dip)
in the way suggested by Tantrum, Murua and Stuetzle (2003) from the
clostest unimodal distribution determined by kernel density estimation
with bandwith chosen so that the density just becomes unimodal. This is
less conservative (and in fact sometimes anti-conservative) than the
values from dip.test.
Usage
dipp.tantrum(xdata,d,M=100)
Arguments
xdata
numeric vector. One-dimensional dataset.
d
numeric. Value of dip statistic.
M
integer. Number of artificial datasets generated in order to
estimate the p-value.
Value
List with components
p.value
approximated p-value.
bw
borderline unimodality bandwith in density
with default settings.
dv
vector of dip statistic values from simulated artificial data.
References
J. A. Hartigan and P. M. Hartigan (1985) The Dip Test of
Unimodality, Annals of Statistics, 13, 70-84.
Tantrum, J., Murua, A. and Stuetzle, W. (2003) Assessment and
Pruning of Hierarchical Model Based Clustering, Proceedings of the
ninth ACM SIGKDD international conference on Knowledge discovery and
data mining, Washington, D.C., 197-205.