Skip to contents

Select Lorenz function

Source: vignettes/review_gd_compute_pip_stats.Rmd
review_gd_compute_pip_stats.Rmd

Background

Povcalnet uses two methods to estimate poverty and inequality statistics from grouped data.

  • One method is based on fitting a Lorenz Quadratic functional form to the grouped data
  • the other one uses a Beta Lorenz function

Povcalnet then choose the statistics to be returned based on a set of rules.

This vignette focuses on the higher level gd_compute_pip_stats() function that handles the application of both functional forms and the selection of the final results.

High level example 

# Input definition
welfare_mean    <- 51.56
ppp             <- 3.69
daily_povline   <- 1.9
monthly_povline <- daily_povline * 365 / 12

# Create grouped data (Type 1)
# http://iresearch.worldbank.org/povcalnet/PovCalculator.aspx
population <- c(0.0005,
                0.0032,
                0.014799999999999999,
                0.0443,
                0.0991,
                0.257,
                0.4385,
                0.5938,
                0.7089,
                1)

welfare <- c(5.824760527229386e-05,
             0.000604029410841011,
             0.0037949334793616948,
             0.013988878652244477,
             0.036992164583098786,
             0.12140708906131342,
             0.24531391873082081,
             0.37446670169288321,
             0.48753116241194566,
             1)


# Estimate poverty statistics
out <- wbpip:::gd_compute_pip_stats(welfare         = welfare,
                                    population      = population,
                                    requested_mean  = welfare_mean,
                                    povline         = monthly_povline,
                                    default_ppp     = ppp)

out
#> $poverty_line
#> [1] 57.79167
#> 
#> $mean
#> [1] 51.56
#> 
#> $median
#> [1] 42.58973
#> 
#> $headcount
#> [1] 0.7184622
#> 
#> $poverty_gap
#> [1] 0.2714275
#> 
#> $poverty_severity
#> [1] 0.1293701
#> 
#> $watts
#> [1] 0.3788881
#> 
#> $gini
#> [1] 0.3123674
#> 
#> $mld
#> [1] 0.1633424
#> 
#> $polarization
#> [1] 0.2556375
#> 
#> $deciles
#>  [1] 0.03750486 0.04963605 0.05871314 0.06775231 0.07741935 0.08831018
#>  [7] 0.10132911 0.11835493 0.14500187 0.25597820