test_fourth: Testing the relationship species attributes and sample...

In weimea: Weighted Mean Analysis

Description

Function calculating fourth corner analysis between sample attributes (environmental variables) and species attributes (traits) and testing it by row based, column based or max permutation test.

Usage

test_fourth(env, com = NULL, traits = NULL, cwm = NULL, perm = 499,
  test = "max", parallel = NULL, p.adjust.method = p.adjust.methods,
  adjustP = FALSE)

## S3 method for class 'testFOURTH'
print(x, digits = max(3, getOption("digits") - 3),
  missing.summary = FALSE, ...)

## S3 method for class 'testFOURTH'
coef(object, ...)

Arguments

env	Vector, matrix or data frame with sample attributes (environemntal variables).
com	Matrix of data frame with species composition data.
traits	Vector, matrix or data frame with species attributes (traits)
cwm	An object of the class cwm.
perm	Number of permutations.
test	Vector of character values. Which test should be conducted? Partial match to 'rowbased' row-based permutation test (called Model 2 in the Legendre et al. 1997 and Dray & Legendre 2008), 'colbased' for column-based permutation test (Model 4), and 'max' for max test (selecting the higher from rowbased and colbased result, Model 6). 'all' includes all three tests. See Details.
parallel	NULL (default) or integer number. Number of cores for parallel calculation of modified permutation test. Maximum number of cores should correspond to number of available cores on the processor.
p.adjust.method	A string indicating the method of P-value adjustement, see p.adjust.methods for possible choices.
adjustP	Logical, default FALSE. Should be the P-values adjusted? If adjustP = TRUE, the last column in the results is adjusted by method selected in p.adjust.method.
x, object	object of the class "cwm", generated by function cwm.
digits	number of digits reported by print method on object of "cwm" class (default is 3).
missing.summary	Logical; should be the summary of values missing in env, cwm and traits be printed along to the result of test_cwm analysis? Default is TRUE.
...	Other arguments for print, summary or coef functions (not implemented yet).

Details

Environmental variables, species composition and traits can be provided in two ways: either each object separately, using arguments env, com and traits, or using only env and cwm; in the later case, the function extract is used to extract the com and traits from the cwm object. Specific issue related to weighted mean is the case of missing species attributes. In current implementation, species with missing species attributes are removed from sample x species matrix prior to permutation of species attributes among species.

Value

Function test_fourth returns the list of the class "testFOURTH" (with print and summary methods), which contains the following items:

• call Function call.

• out Data frame with results of calculation; first two columns contain the original fourth corner metric (r_fc) and its Chessel transformation (r_ch); following columns contain some of the following results of permutation tests: P.row, P.col and P.max. If adjustP = TRUE, the last column contains P values adjusted by p.adjust function.

• miss List with missing values of env, traits or cwm (the last provided for compatibility with test_cwm function and refers to samples which, after removing missing traits values, do not contain any species.)

• param Setting of input parameters (perm and test).

See Also

cwm test_cwm

Examples

data (vltava)
vltava.env <- vltava$env[,c('pH', 'COVERE32')]
vltava.com <- vltava$herbs$spe
vltava.traits <- vltava$herbs$traits
fc <- test_fourth (env = vltava.env, com = vltava.com, traits = vltava.traits)

weimea documentation built on May 2, 2019, 4:44 p.m.