iGini: Individual decomposition of the Gini Index

In iIneq: Computing Individual Components of the Gini and the Theil Indices

Description

The function computes individual components of the Gini index and their group-based decompositions. It takes as input an outcome variable, a grouping variable, and an optional sampling weight. It returns a data matrix of three columns containg individual contributions and their between- and within-group components. Because iGini is computational insensive, parallel processing is recommended, and the number of cores can be specified.

Usage

iGini(x, g, w=rep(1,length(x)),core=1)

Arguments

x	Input continuous variable such as income.
g	A grouping variable containing integers, such gender coded 1 & 2.
w	An optional sampling weight variable.
core	An optional input for specifying the number of processing cores in your computer. When specified, you will need to have the doParallel package and the foreach package installed for conducting parallel processing to speed up the computation.

Value

The function outputs three variables, g.i, g.ikb, and g.ikw.

g.i	This variable gives the individual contribtions to the overall Gini index.
g.ikb	This variable provides for each individual component of the Gini its between-group subcomponent.
g.ikw	This variable provides for each individual component of the Gini its within-group subcomponent. The g.ikb and g.ikw sum up to g.i for each i observation.

References

Tim F. Liao. 2019. "Individual Components of Three Inequality Measures for Analyzing Shapes of Inequality.” Sociological Methods & Research Advance online publication. doi:10.1177/0049124119875961

Examples

data(ChickWeight)
attach(ChickWeight)
iGini.result <- iGini(weight,Diet,core=1)

iIneq documentation built on Jan. 21, 2021, 1:07 a.m.