nn.ent: Works out entropy of a sample.

In dennisprangle/abctools: Tools for ABC Analyses

Works out entropy of a sample.

Description

The function computes the k nearest neighbour sample entropy.

Usage

nn.ent(th, k=4)

Arguments

th	The sample from which to compute the entropy.
k	The order (number of neighbours) of the sample entropy calculation.

Details

The sample entropy gives a measure of information in a (posterior) sample, or lack of it.

Value

The k nearest neighbour entropy from the sample.

Warning

For high-dimensional posterior samples, the nn.ent calculation is quite computationally intensive.

Author(s)

Matt Nunes

References

Nunes, M. A. and Prangle, D. (2016) abctools: an R package for tuning approximate Bayesian computation analyses. The R Journal 7, Issue 2, 189–205.

Singh, H. et al. (2003) Nearest neighbor estimates of entropy. Am. J. Math. Man. Sci.,23, 301–321.

Shannon, C. E. and Weaver, W. (1948) A mathematical theory of communication. Bell Syst. Tech. J., 27, 379–423.

See Also

mincrit

Examples

# create a dummy sample to calculate an entropy measure:

theta<-rnorm(10000)

nn.ent(theta)

dennisprangle/abctools documentation built on Sept. 22, 2023, 9:50 p.m.