count: OPHI counting approach

In parsec: PARtial orders in Socio-EConomics

Description

The function implements in a single call the computations involved in OPHI counting approach. The OPHI approach can be applied also when cardinal variables are of concern. The function implements the methodology for ordinal variables only.

Usage

count(y, ...)
## Default S3 method:
count(y, z, w=rep(1, ncol(y)), k=sum(w), freq=rep(1, nrow(y)), ...)
## S3 method for class 'wprof'
count(y, ...)

Arguments

y	matrix of profiles, it can be substituted by an object of class wprof.
z	vector of cutoffs.
w	variables weights.
k	overall cutoff.
freq	profiles frequencies, if y is an object of class wprof this argument can be omitted.
...	any of the above.

Value

An object of S3 class ophi that contains all the informations and evaluations related to the OPHI counting approach. The object is a list of the following arguments

y	the matrix of profiles,
freq	the profiles frequencies,
d	the variables number,
n	the observations number (frequencies sum),
z	vector of cutoffs,
k	overall cutoff,
rho	the function used to compare the profiles with the vector of cutoffs,
rho_k	the function used to compare the profiles with the overall cutoff by weighting the variables,
g0	a matrix profiles-variables reporting the results of the function rho,
c	the censored vector of deprivation counts,
Z_k	a boolean vector that for each profile indicate if it identifies a state of deprivation,
q	the number of poor in the population of n units,
H	the headcount ratio,
A	the average deprivation share ,
M0	the adjusted headcount ratio.

References

Alkire S., Foster J. (2011), Counting and multidimensional poverty measurement, Journal of Public Economics, 96(7-8), 476-487.

Examples

vl <- c(2, 3, 3, 2)
prof <- var2prof(varlen = vl)

res <- count(prof, z = c(1, 2, 1, 1), k = 1)

res

parsec documentation built on May 2, 2019, 6:08 p.m.