ChaoEntropy: Estimation of Shannon entropy

In JohnsonHsieh/ChaoEntropy: Statistical package 'ChaoEntropy'

Description

ChaoEntropy is a function to provide a new Shannon entropy estimator proposed by Chao et al. (2013).

For individual-based (abundance) data, there are five other estimators which are Chao and Shen (2003) estimator; Grassberger (2003) estimator; Zhal (1977) jackknife estimator; Zhang (2012) Hz* estimator; and the observed provided in comparison.

For sample-based (incidence) data, new Shannon entropy estimator is in comparison with the observed entropy.

Usage

ChaoEntropy(data, datatype = c("abundance", "incidence"), method = c("all", "Chao",
            "ChaoShen", "Grassberger", "Jackknife", "Zhang", "Observed"),
            se = TRUE, nboot = 200, conf = 0.95)

Arguments

data	a vector of species abundance or incidence frequency. If datatype = "incidence", then the input format of first entry should be total number of sampling units, and followed by species incidence frequency.
datatype	the data type of input data. That is individual-based abundance data (datatype = "abundance") or sample-based incidence data (datatype = "incidence").
method	the method constructed to estimate entropy (see Details)
se	calculate bootstrap standard error and show confidence interval; default is TRUE.
nboot	the number of bootstrap resampling times, default is 200.
conf	a positive number ≤ 1. "conf" specifies the confidence level for confidence interval. The default is 0.95.

Details

1. If datatype = "abundance":

   • method = "all" : all estimators below involved. The default is "all"

   • method = "Chao" estimator, see Chao et al. (2013)

   • method = "ChaoShen" estimator, see Chao and Shen (2003)

   • method = "Grassberger" estimator, see Grassberger (2003)

   • method = "Jackknife" estimator, see Zhal (1977)

   • method = "Zhang" estimator, see Zhang (2012)

   • method = "Observed" estimator, the observed entropy estimator

2. If datatype = "incidence":

   • method = "all" : all estimators below involved. The default is "all"

   • method = "Chao" estimator, see Chao et al. (2013) in Appendix S6

   • method = "Observed" estimator, the observed entropy estimator

Value

ChaoEntropy returns a table of various entropy estimators, their standard error and 95 % confidence interval which the method you choose.

Author(s)

Y.T. Wang ; Y.H. Lee ; K.S. Tseng ; Anne Chao
Maintainer : Anne Chao [e-mail]

References

Chao, A., Wang, Y.T. & Jost, L. (2013) Entropy and the species accumulation curve: a novel entropy estimator via discovery rates of new species. To appear in Methods in Ecology and Evolution.

Chao, A. & Jost, L. (2012) Coverage-based rarefaction and extrapolation: standardizing samples by completeness rather than size. Ecology, 93, 2533-2547.

Chao, A. & Shen, T.J. (2003) Nonparametric estimation of Shannon's index of diversity when there are unseen species. Environmental and Ecological Statistics, 10, 429-443.

Grassberger, P. (2003) Entropy estimates from insufficient samplings. URL www.arxiv.org. arXiv:physics/0307138v2. Updated 2008.

Zahl, S. (1977) Jackknifing an index of diversity. Ecology, 58, 907-913.

Zhang, Z. (2012) Entropy estimation in Turing's perspective. Neural Computation, 24, 1368-1389.

Chao, A. (1984) Nonparametric estimation of the number of classes in a population. Scandinavian Journal of Statistics, 11, 265-270.

Chao, A. (1987) Estimating the population size for capture-recapture data with unequal catchability. Biometrics, 43, 783-791.

See Also

Count2Abun, Count2Inci, Matrix2Inci

Examples

# load the individual-base (abundacne) data
data(Spider_data)

# Estimation of Shannon entropy
ChaoEntropy(Spider_data, datatype="abundance", method="all", se=TRUE, nboot=200, conf=0.95)

# load the sample-base (incidence) data
data(Ant_data)

# Estimation of Shannon entropy
ChaoEntropy(Ant_data, datatype="incidence", method="all", se=TRUE, nboot=200, conf=0.95)

JohnsonHsieh/ChaoEntropy documentation built on May 7, 2019, 12:01 p.m.