MigConnectivity (version 0.4.7)

diffMC: Pairwise differences between two or more independent MC estimates

Description

Estimates mean (and median) differences in MC, and includes measures of uncertainty (SE and CI). For those measures of uncertainty to be accurate, only apply this function to MC estimates where all data sources are independent (e.g., different species).

Usage

diffMC(estimates, nSamples = 1e+05, alpha = 0.05, returnSamples = FALSE)
diffStrength(estimates, nSamples = 1e+05, alpha = 0.05, returnSamples = FALSE)

Value

diffMC returns a list with elements:

meanDiff, medianDiff

Vectors with mean and medians of sampled differences for each pairwise comparison. Estimates of difference between MC values incorporating parametric uncertainty.

seDiff

Vector with standard errors of MC differences for each pairwise comparison, estimated from SD of sampled differences.

simpleCI

Matrix of 1 - alpha confidence intervals for MC differences, estimated as alpha/2 and 1 - alpha/2 quantiles of sampleMC.

bcCI

Matrix of bias-corrected 1 - alpha confidence intervals for MC differences for each pairwise comparison. Preferable to simpleCI when meanDiff is the best estimate of the MC difference. simpleCI is preferred when medianDiff is a better estimator. When meanDiff==medianDiff, these should be identical. Estimated as the pnorm(2 * z0 + qnorm(alpha / 2)) and pnorm(2 * z0 + qnorm(1 - alpha / 2)) quantiles of sampled differences, where z0 is the proportion of sampleDiff < meanDiff.

sampleDiff

Only provided if returnSamples is TRUE. List of sampled values for each pairwise MC difference.

Arguments

estimates

List of at least two MC estimates, provided by the estMC function. If this is a named list (recommended), the function will use these names in labeling the differences.

nSamples

A positive integer, number of samples (with replacement) to draw from each pair of MC estimates (default 100000). If set to NULL, compares all MC samples from each pair.

alpha

Level for confidence/credible intervals provided.

returnSamples

Should the function return all the sampled differences? Defaults to FALSE to reduce storage requirements. Change to TRUE to compute your own summary statistics.

References

Cohen, E. B., C. S. Rushing, F. R. Moore, M. T. Hallworth, J. A. Hostetler, M. Gutierrez Ramirez, and P. P. Marra. 2019. The strength of migratory connectivity for birds en route to breeding through the Gulf of Mexico.

Examples

# \donttest{
data('OVENdata')
ovenPsi <- estTransition(isGL = OVENdata$isGL, #Logical vector:light-level GL(T)
                 isTelemetry = !OVENdata$isGL,
                 geoBias = OVENdata$geo.bias, # Light-level GL location bias
                 geoVCov = OVENdata$geo.vcov, # Location covariance matrix
                 targetSites = OVENdata$targetSites, # Non-breeding target sites
                 originSites = OVENdata$originSites, # Breeding origin sites
                 originPoints = OVENdata$originPoints, # Capture Locations
                 targetPoints = OVENdata$targetPoints, # Device target locations
                 verbose = 0,   # output options
                 nSamples = 100, # This is set low for example
                 resampleProjection = sf::st_crs(OVENdata$targetSites))
ovenEst <- estStrength(targetDist = OVENdata$targetDist, # targetSites distance matrix
                 originDist = OVENdata$originDist, # originSites distance matrix
                 originRelAbund = OVENdata$originRelAbund,#Origin relative abund
                 psi = ovenPsi,
                 verbose = 1,   # output options
                 nSamples = 1000)
fm <- getCMRexample()
originPos13 <- matrix(c(rep(seq(-99, -81, 2), each = 10),
                        rep(seq(49, 31, -2), 10)), 100, 2)
targetPos13 <- matrix(c(rep(seq(-79, -61, 2), each = 10),
                        rep(seq(9, -9, -2), 10)), 100, 2)
originPosCMR <- rowsum(originPos13, c(rep(1:2, 5, each = 5),
                                      rep(3:4, 5, each = 5))) / 25
targetPosCMR <- rowsum(targetPos13, c(rep(1:2, 5, each = 5),
                                      rep(3:4, 5, each = 5))) / 25
originDist <- distFromPos(originPosCMR, 'ellipsoid')
targetDist <- distFromPos(targetPosCMR, 'ellipsoid')
originRelAbundTrue <- rep(0.25, 4)
theorEst <- estStrength(originRelAbund = originRelAbundTrue, psi = fm,
                  originDist = originDist, targetDist = targetDist,
                  originSites = 5:8, targetSites = c(3,2,1,4),
                  nSamples = 1000, verbose = 0,
                  sampleSize = length(grep("[2-5]", fm$data$data$ch)))
ovenEst
theorEst
diff1 <- diffMC(estimates = list(Ovenbird = ovenEst, Theorybird = theorEst),
                nSamples = 10000, returnSamples = TRUE)

# }