Variance partition and WIC/TNW

Description

Variance partition based on a single multi-level factor and comparison of the relative contribution of the “Between Group Component” (BGC) versus the ratio WIC/TNW. For a more general discussion of variance partition the reader is pointed to Viola et al. (2012).

Usage

1
Hier2L(dataset, factor = 1, weight.type = "N_items")

Arguments

dataset

Object of class RInSp with data of type “double”.

factor

Numeric value indicating the column in the info part of the RInSp input dataset containing the factor used for build subsets.

weight.type

The weighting option for unbalanced prey numbers. See details in WTcMC.

Details

The Beteen Group Component is calculated as the (population) variance of the mean values of the n-levels groups.

The level of checking performed by the procedure is low. Pay attention to input parameters.

Value

The result is a matrix of 5 rows by n+1 columns, when n is the number of levels of the factor used to create groups in the dataset. The first four rows are the values for WIC, BIC, TNC and WIC/TNW for all possible levels and the whole dataset. Last row is the value of the “Between Group Component” (BGC) for the n-levels factor.

Author(s)

Dr. Nicola ZACCARELLI and Dr. Giorgio MANCINELLI

References

Violle C., Enquist B.J., McGill B.J., Jiang L., Albert C.H., Hulshof C., Jung V. and Messier J. 2012. The return of the variance: intraspecific variability in community ecology. Trends in Ecology & Evolution 27: 244252.

See Also

Function WTcMC.

Examples

1
2
3
4
5
# Example of variance decomposition
data(Trout)
TroutRIS= import.RInSp(Trout, col.header=TRUE, row.names=1, info.cols=2, data.type="double")
decomp = Hier2L(TroutRIS, factor=1)
rm(list=ls(all=TRUE))