Skip to contents

Load the bundled BIDIRECTIONAL 158 <-> 140 neighbourhood crosswalk

Source: R/toronto_neighbourhoods.R
morie_to_hood_crosswalk.Rd

Returns the bundled inst/extdata/to_hood_158_140_crosswalk.csv, computed from polygon intersection of the two upstream Open Toronto GeoJSON layers (Neighbourhoods - 4326.geojson and Neighbourhoods - historical 140 - 4326.geojson) reprojected to EPSG:3347 (NAD83 Statistics Canada Lambert – metric, accurate areas). Seven columns:

Usage

morie_to_hood_crosswalk()

Value

A data.frame with the columns above. hood_140 and hood_158 are character (zero-padded to 3 chars).

Details

hood_140

3-char zero-padded historical short code

name_140

historical neighbourhood name (carries "(NN)" suffix)

hood_158

3-char zero-padded current short code

name_158

current neighbourhood name

pct_140_in_158

FORWARD: percent of the 140's area inside this 158. Per-140 rows sum to 100. Used by morie_tps_disaggregate_140_to_158() as the cake-cutting weight under a uniform-density assumption.

pct_158_in_140

REVERSE: percent of the 158's area inside this 140. Per-158 rows sum to 100. For pure cake-cuts every split child has pct_158_in_140 == 100 (each 158 is entirely inside its parent 140), so morie_tps_aggregate_158_to_140() is mathematically EXACT (lossless sum) for the 1:1 + split cohort. Only the one split+merge sliver in the bundled OT data has a non-100 reverse percent.

relation

"1:1" / "split" (one 140 -> N 158s) / "merge" (multiple 140s -> one 158) / "split+merge"

Empirical distribution on the bundled OT data:

  • 123 1:1 rows (78\

  • 34 split rows (16 historical hoods) – pct_158_in_140 == 100

  • 1 merge – both percents == 100

  • 1 split+merge – one sliver < 100