Bivariate Moran's I between two TPS categories — morie_tps_bivariate_morans

Tests whether category A's per-neighbourhood count co-varies with category B's count in NEIGHBOURING neighbourhoods (spatial spillover). Builds a k-NN row-standardised spatial weights matrix from per-hood centroids derived from category A's WGS84 latitude/longitude. Reports Pearson r alongside as a non-spatial baseline.

Usage

morie_tps_bivariate_morans_i(dfs, cat_a, cat_b, k_neighbours = 5L)

Arguments

dfs: Named list of TPS data.frames keyed by category.
cat_a: Name of category A in dfs.
cat_b: Name of category B in dfs.
k_neighbours: Number of nearest neighbours per row in W (default 5L).

Value

A morie_tps_result named list.

Examples

set.seed(1)
df <- data.frame(OCC_YEAR = rep(2014:2023, each = 30),
                 OCC_MONTH = sample(month.name, 300, TRUE),
                 HOOD_158 = sample(sprintf("%03d", 1:20), 300, TRUE),
                 LAT_WGS84 = runif(300, 43.6, 43.8),
                 LONG_WGS84 = runif(300, -79.5, -79.2))
dfs <- list(a = df, b = df)
res <- try(morie_tps_bivariate_morans_i(dfs, "a", "b"))
if (!inherits(res, "try-error")) str(res, max.level = 1)
#> List of 8
#>  $ title         : chr "Bivariate Moran's I -- a vs b"
#>  $ call          : chr "morie_tps_bivariate_morans_i(dfs, ‘a’, ‘b’)"
#>  $ summary_lines :List of 6
#>  $ tables        : list()
#>  $ warnings      : chr(0) 
#>  $ interpretation: chr "I_AB=+0.055, Pearson r=+1.000.  Positive I_AB means a counts in a hood track b counts in NEIGHBOURING hoods (sp"| __truncated__
#>  $ payload       :List of 7
#>  $ sections      : list()
#>  - attr(*, "class")= chr [1:3] "morie_tps_result" "morie_rich_result" "list"