unbal.rep.CI: Measure of Repeatability for Diet Estimates...

In QFASA: Quantitative Fatty Acid Signature Analysis

Measure of Repeatability for Diet Estimates (Unbalanced/Missing Value Case)

Description

Computes a measure of repeatability for a sample of predators with repeated diet estimate measurements (Unbalanced/Missing Value Case)

Usage

unbal.rep.CI(
  data,
  B,
  R,
  CI = TRUE,
  alpha,
  prey.database,
  fatcont.mat,
  dist.meas,
  ext.fa,
  gamma.QFASA = 1,
  gamma.rho = 1
)

Arguments

data	data frame of diet estimates. First column must denote the predator and second column the second factor (eg. year or season). Need not contain the same number of repeated measurements for each predator.
B	number of pseudo predators samples to generate for bias calculation. Default is set to 50 because is slow to run.
R	number of bootstrap samples (i.e. R samples for each generated sample of pseudo predators). Default is set to 100 because it is slow to run.
CI	indicates if a confidence interval for rho is to be calculated. Default is FALSE since this is time consuming to obtain.
alpha	a (1-alpha/2)X100 percent confidence interval is calculated for rho if CI=TRUE.
prey.database	prey data base that was used to compute the QFASA diet estimates in data. Will be used to generate pseudo predators.
fatcont.mat	data frame or matrix of length equal to the number of prey FA signatures in prey data base. First column is name of species and second column is lipid.
dist.meas	distance measure to use in p.QFASA.
ext.fa	subset of FAs to use.
gamma.QFASA	if dist.meas=3, gamma is required. Default is 1.
gamma.rho	value of gamma to be used to compute CS distance in repeatablity functions. Default is 1.

Value

Bias corrected measure of repeatability, estimate of the bias and (if CI=TRUE) a confidence interval for the true repeatability.

QFASA documentation built on Sept. 11, 2024, 8:08 p.m.