Skip to contents

Simulated records with a known disparity

Source: R/fairsim.R
fairsim.Rd

Incidents in segregated areas, whose true occurrence is the same for every group but whose recording, and the risk an algorithm predicts for them, differ by group by a set amount: data on which an audit's answer is known (after Barman and Barman 2026). The first group is the reference.

Usage

fairsim(
  n = 2000,
  groups = c("A", "B"),
  props = NULL,
  nareas = 20,
  base = 0.3,
  bias = 0.5,
  seed = NULL
)

Arguments

n

The number of records: a whole number from 1.

groups

The group labels, at least two, the reference first.

props

NULL for equal shares, or one weight from 0 per group (not all 0).

nareas

The number of areas: a whole number, at least the number of groups.

base

The true rate, and the reference group's recorded rate: one number from 0 to 1.

bias

The disparity injected: one number from -1 to 1.

seed

NULL to draw the seed from R's generator, or one whole number from 0; the records come from crimlab's own generator, the same on every platform.

Value

A data frame with one row per record: area (a number from 1), group, truth, det and risk.

Details

Each record draws its group by props, then an area among those its group lives in (with g groups, areas 1, g + 1, 2 g + 1, ... hold the first group only, areas 2, g + 2, ... the second, and so on), then truth, 1 with probability base whatever the group; det, the recorded outcome, 1 with probability base in the reference group and base (1 - bias) in the others (clamped to 0 and 1), so that the other groups' incidents are recorded less when bias is positive; and risk, normal with mean 250 in the reference group and 250 + 100 bias in the others and standard deviation 40, clamped to 0 and 500. So fairdf() on det should find a ratio near 1 - bias, scrdsp() on risk a gap near 100 bias, and calaud() the other groups' areas ranked as riskier than their recorded outcomes.

References

Barman, P. K. and Barman, P. K. (2026). Spatial disparities in simulated patrol detection: a reproducible generative-model audit across Baltimore and Chicago. doi:10.48550/arXiv.2603.18987

Examples

d <- fairsim(2000, seed = 1)
head(d)
#>   area group truth det     risk
#> 1   10     B     0   0 379.1753
#> 2    2     B     0   1 324.0614
#> 3   10     B     0   1 280.4492
#> 4    1     A     0   0 312.1302
#> 5    6     B     1   0 326.8925
#> 6    8     B     0   1 304.4651
# the injected ratio is 1 - 0.5
fairdf(d$det, d$group, base = "A")$tab[, c("group", "di", "dilo", "dihi")]
#>   group        di      dilo      dihi
#> 1     A 1.0000000       NaN       NaN
#> 2     B 0.5751675 0.4849938 0.6821069