Skip to contents

Per-individual segregation-placement-count concentration on OTIS b09

Source: R/mrm_otis.R
mrm_otis_placement_concentration.Rd

Expands the OTIS b09 banded per-individual placement counts into a per-person vector using band midpoints (the published bands are \{1, 2, 3, 4, 5, 6-10, 11-15, 16-20, 21-25, 26-30, 31-35, 36-40, >40\}), then computes Hill-MLE Pareto exponent, Gini coefficient, and top-k% concentration within each fiscal year and pooled.

Usage

mrm_otis_placement_concentration(
  data,
  year_col = "EndFiscalYear",
  band_col = "NumberPlacements_Segregation",
  count_col = "NumberIndividuals_Segregation",
  gender_col = NULL,
  gender_keep = NULL,
  x_min = 1L,
  top_pct = 0.05
)

Arguments

data

A data.frame in b09 long format with the columns named in year_col, count_col, band_col, optionally gender_col.

year_col

Column name of the fiscal-year identifier (default "EndFiscalYear").

band_col

Column name of the placement-count band (default "NumberPlacements_Segregation").

count_col

Column name of the per-band individual count (default "NumberIndividuals_Segregation").

gender_col

Optional gender filter column. If supplied with gender_keep, rows are restricted to the kept genders.

gender_keep

Character vector of gender values to retain.

x_min

Hill-MLE lower-tail cutoff (default 1L).

top_pct

Numeric in (0, 1); top concentration cutoff (default 0.05).

Value

A data.frame with one row per fiscal year plus a final "pooled" row, containing columns year, n_individuals, n_placements, mean_per_individual, gini, hill_alpha, top_pct_share.

References

Hill, B. M. (1975). A simple general approach to inference about the tail of a distribution. The Annals of Statistics, 3(5), 1163-1174.

Clauset, A., Shalizi, C. R., & Newman, M. E. J. (2009). Power-law distributions in empirical data. SIAM Review, 51(4), 661-703.

Examples

if (FALSE) {
  b09 <- read.csv("b09_individuals_in_segregation_number_of_times_in_segregation.csv")
  mrm_otis_placement_concentration(b09)
}