acid (version 1.1)

ineq.md: Three Inequality Measures for a Mixture of Income Distributions

Description

This function uses Monte Carlo methods to estimate an the mean logarithmic deviation, the Theil Index and the Gini Coefficient for a mixture of two continuous income distributions and a point mass for zero-incomes.

Usage

ineq.md(n, dist1, dist2, theta, 
p0, p1, p2, dist.para.table, zero.approx)

Arguments

n

sample size used to estimate the gini coefficient.

dist1

character string with the name of the first continuous distribution used. Must be listed in dist.para.table. Must be equivalent to the respective function of that distribution, e.g. norm for the normal distribution.

dist2

character string with the name of the second continuous distribution used. Must be listed in dist.para.table. Must be equivalent to the respective function of that distribution, e.g. norm for the normal distribution.

theta

vector with the parameters of dist1 and dist2. Order must be the same as in the functions for the distributions.

p0

scalar with probability mass for the point mass.

p1

scalar with probability mass for dist1.

p2

scalar with probability mass for dist2.

dist.para.table

a table of the same form as dist.para.t with distribution name, function name and number of parameters.

zero.approx

a scalar which replaces zero-incomes (and negative incomes), such that entropy measures involving a logarithm return finite values.

Value

MLD

the estimated mean logarithmic deviation.

Theil

the estimated Theil index.

Gini

the estimated Gini coefficient.

y

a vector with the simulated incomes to estimate the entropy measure.

y2

a vector with the zero-replaced simulated incomes to estimate the entropy measure.

zero.replace

a logical vector indicating whether a zero has been replaced.

stat

a vector with the simulated group the observation was chosen from. 0 is the point mass, 1 dist1 and 2 dist2.

References

Cowell, F.A. (2000): Measurement of Inequality, in: Atkinson and Bourguignon (eds.), Handbook of Income Distribution, pp. 87-166, Elsevier, Amsterdam.

See Also

dist.para.t, gini, entropy

Examples

 
theta<-c(0,1,5,2)
x<- c(rgamma(500,2,1),rgamma(500,5,2))
entropy(x,0)
entropy(x,1)
gini(x)$Gini
data(dist.para.t)
im<-ineq.md(100,"gamma","gamma",theta,0,0.5,0.5,dist.para.t,zero.approx=1)
im$MLD
im$Theil
im$Gini