metricTester (version 1.0.0)

summCorrs: Summarize correlations among metrics over a result from a varyX function

Description

Takes the results of one of the varyX functions, and calculates the correlations among metrics, returning either the raw or summarized correlations.

Usage

summCorrs(vary.results, exclude, return.raw = FALSE,
  cor.method = "spearman")

Arguments

vary.results

Results from any of the varyX (e.g., varyAbundance) functions.

exclude

Results columns to exclude from correlation. For instance, with alpha metrics, one would want to, at the minimum, exclude the plot name column.

return.raw

Default is FALSE. Whether to return the raw correlation coefficients between the metrics for each element from vary.results, or whether to summarize the correlations by their mean per parameter set from vary.results.

cor.method

Default is "spearman", but takes any of the other options from the base cor function.

Value

• If return.raw is set to FALSE, returns a list of matrices, one for each set of parameters in vary.results, summarizing the mean correlation coefficients between each pairwise metric correlation. If return.raw is set to TRUE, returns a list of lists, one for each set of parameters. Each of these lists is the length of the number of iterations in vary.results. Each element in the list is a matrix of pairwise metric correlations.

Details

Not a well tested function.

References

Miller, E. T., D. R. Farine, and C. H. Trisos. 2015. Phylogenetic community structure metrics and null models: a review with new methods and software. bioRxiv 025726.

Examples

system.time(vSize <- varyX(alpha=TRUE, tree.size=c(59, 100), richness=40:59, delta=1,
	abundances=round(rlnorm(5000, meanlog=2, sdlog=1)) + 1, iterations=2, cores=1))

test <- summCorrs(vSize, exclude=c("plot", "richness"))