bias.all: Calculate bias correction for confidence intervals from...

In QFASA: Quantitative Fatty Acid Signature Analysis

Calculate bias correction for confidence intervals from beta.meths.CI.

Description

Calculate bias correction for confidence intervals from beta.meths.CI.

Usage

bias.all(
  p.mat,
  prey.mat,
  cal.mat = rep(1, length(ext.fa)),
  fat.cont = rep(1, nrow(prey.mat)),
  R.bias,
  noise,
  nprey,
  specify.noise,
  dist.meas,
  ext.fa
)

Arguments

p.mat	matrix containing the fatty acid signatures of the predators.
prey.mat	matrix containing a representative fatty acid signature
cal.mat	matrix of calibration factors where the i th column is to be used with the i th predator. If modelling is to be done without calibration coefficients, simply pass a vector or matrix of ones.
fat.cont	prey fat content
R.bias	bootstrap replicates
noise	noise
nprey	number of prey
specify.noise	noise
dist.meas	distance measure
ext.fa	subset of FA's to use.

Value

Row 1 is Lambda CI, row 2 is Lambda skew, and row 3 is Beta CI

Examples

## Fatty Acids
data(FAset)
fa.set = as.vector(unlist(FAset))

## Predators
data(predatorFAs)
tombstone.info = predatorFAs[,1:4]
predator.matrix = predatorFAs[, fa.set]
npredators = nrow(predator.matrix)

## Prey
prey.sub = preyFAs[, fa.set]
prey.sub = prey.sub / apply(prey.sub, 1, sum)
group = as.vector(preyFAs$Species)
prey.matrix.full = cbind(group,prey.sub)
prey.matrix = MEANmeth(prey.matrix.full)

## Calibration Coefficients
data(CC)
cal.vec = CC[CC$FA %in% fa.set, 2]
cal.mat = replicate(npredators, cal.vec)

# Note: uncomment examples to run. CRAN tests fail because execution time > 5 seconds
# diet.est <- p.QFASA(predator.mat = predator.matrix,
#                     prey.mat = prey.matrix,
#                     cal.mat = cal.mat,
#                     dist.meas = 2,
#                     start.val = rep(1,nrow(prey.matrix)),
#                     ext.fa = fa.set)[['Diet Estimates']]
#
# bias <- bias.all(p.mat = diet.est,
#                  prey.mat = prey.matrix.full,
#                  cal.mat = cal.mat,
#                  R.bias = 10,
#                  noise = 0,
#                  nprey = 50,
#                  dist.meas = 2,
#                  ext.fa = fa.set)

QFASA documentation built on Nov. 17, 2023, 1:08 a.m.