| Pred.incidence.rare {RSE} | R Documentation |
Incidence-based data: predicting the number of new rare species
Description
Incidence-based prediction on the number of new rare species using a Bayesian-weight and two unweighted estimators along with their bootstrap standard errors and 95% bootstrap confidence intervals.
Usage
Pred.incidence.rare(boot.rep = 100, Q = NULL, xi = NULL, nT, u, k.show = 3)
Arguments
boot.rep |
Replicate number of the bootstrapping procedure |
Q |
A vector of species frequency counts, i.e., the number of species dectected once (in only one quadrat), the number of species dectected twice (in exactly two quadrats), and so forth. |
xi |
A vector of species incidence counts, i.e., the number of quadrats with species 1, the number of quadrats with species 2, and so forth. |
nT |
The number of quadrats of the original sample |
u |
The number of quadrats of an additional sample |
k.show |
Display the estimating results of the numbers of new rare species detected in the number of quadrats <= k.show in the additional sample |
Value
Estimating results including point estimate, bootstrap standard error, and 95 % bootstrap confidence interval for each of three methods (a Bayesian-weight and two unweighted estimators)
Author(s)
Youhua Chen & Tsung-Jen Shen
References
Shen TJ, Chen YH (2018) A Bayesian weighted approach to predicting the number of newly discovered rare species. Conservation Biology, In press.
See Also
Examples
## As an example, Canadian-mite data are used here.
data(CanadaMite)
## two columns represent two samples of incidence counts
X.merge = CanadaMite
## the first column is treated as the original sample
X.col1 = X.merge[,1]
Xi = X.col1
## Convert species incidence count data to frequency counts data
Q = X.to.f(Xi)
## the number of quadrats in the first sample
nT = 16
## the number of quadrats in the additional sample (i.e., the second column)
u = 16
Pred.incidence.rare(Q=Q, nT=nT, u=u)