Aggregate per-record predictive-policing data to one row per area

Usage

morie_predpol_aggregate_areas(
  area,
  risk,
  outcome,
  group = NULL,
  population = NULL
)

Arguments

area: Area identifier for each record.
risk: Predicted risk score for each record.
outcome: Realised-outcome indicator/count for each record.
group: Optional protected attribute per record; the per-area majority value becomes that area's group label.
population: Optional area population: a named numeric vector (area -> population) or a per-record vector. When given, the outcome rate is per 10,000 inhabitants; otherwise it is the mean outcome per record.

Value

A named list: areas, mean_risk, outcome_rate, group, n_records.

Examples

agg <- morie_predpol_aggregate_areas(
  area = c("a", "a", "b", "b"), risk = c(10, 20, 30, 40),
  outcome = c(1, 0, 1, 1)
)
agg$mean_risk # 15 35
#> [1] 15 35
agg$outcome_rate # 0.5 1.0
#> [1] 0.5 1.0