build_zones: Build Candidate Spatial Zones Using Circular Windows
In treeSS: Tree-Spatial Scan Statistic for Cluster Detection

build_zones

R Documentation

Build Candidate Spatial Zones Using Circular Windows

Description

Constructs the set of candidate spatial zones for the circular scan statistic. Each zone is defined by a circular window centered on a region's centroid, containing all regions whose centroids fall within the circle.

Usage

build_zones(regions, max_pop = NULL)

Arguments

`regions`	A `data.frame` with columns `region_id`, `population`, `x`, and `y`. Coordinates `x` and `y` represent centroids of each region.
`max_pop`	A numeric value specifying the maximum population allowed inside a zone. If `NULL` (default), it is set to 50% of the total population.

Details

For each region j, regions are sorted by distance from j. Regions are added incrementally to the zone as long as the total population does not exceed max_pop. This generates a nested set of circular windows centered on each region.

Value

A list of zones. Each zone is a list with elements:

center: Integer index of the center region.
region_idx: Integer vector of region indices inside the zone.
population: Total population inside the zone.

References

Kulldorff, M. (1997). A spatial scan statistic. Communications in Statistics - Theory and Methods, 26(6), 1481–1496.

Examples

regions <- data.frame(
  region_id = 1:5,
  population = c(1000, 2000, 1500, 800, 1200),
  x = c(0, 1, 2, 0.5, 1.5),
  y = c(0, 0, 0, 1, 1)
)
zones <- build_zones(regions, max_pop = 3000)
length(zones) # number of candidate zones

treeSS documentation built on May 16, 2026, 1:08 a.m.