Skip to contents

HealthConcentrationIndex: \(2\,cov(h, r)/\bar h\) with fractional ranks (mean ranks for ties) and the Erreygers correction (Kakwani, Wagstaff and van Doorslaer 1997). SpatialGini: Gini as ineq::Gini with the Rey and Smith (2013) neighbour / non-neighbour decomposition. TheilDecomposition: Theil T and L (ineq::Theil) with the between/within decomposition (Shorrocks 1980). Identical to the Python arm morie.fn.hlthacc.

Usage

HealthConcentrationIndex(health, rank_variable)

SpatialGini(x, W = NULL)

TheilDecomposition(x, groups = NULL)

Arguments

health

Health variable.

rank_variable

Ranking variable.

x

Values.

W

Optional weights (nonzero = neighbour).

groups

Optional group labels.

Value

List.

References

Kakwani, N., Wagstaff, A. and van Doorslaer, E. (1997). Socioeconomic inequalities in health: measurement, computation, and statistical inference. Journal of Econometrics 77, 87-103.

Rey, S. J. and Smith, R. J. (2013). A spatial decomposition of the Gini coefficient. Letters in Spatial and Resource Sciences 6, 55-70.

Examples

HealthConcentrationIndex(c(4, 3, 2, 1), c(10, 20, 30, 40))$index
#> [1] -0.25
SpatialGini(1:4)$gini
#> [1] 0.25
TheilDecomposition(1:4)$T
#> [1] 0.1064401