frontier_detect: Method for detecting frontiers

In MengLeZhang/socialFrontiers: What the Package Does (One Line, Title Case)

Method for detecting frontiers

Description

'frontier_detect' returns a list object containing a adjacency matrix with 0 if the border between 2 areal units is a frontiers and 1 if not. NA if two units are not bordering. Currently only implements the algorithm for frontier detection from Dean et al

Usage

frontier_detect(
  y,
  data,
  n.trials,
  W.nb = NULL,
  fix.rho = TRUE,
  rho = 0.99,
  ...
)

Arguments

y	A string indicating column in data denoting the value to find discontinuities in
data	A sp or sf file of areal units from which we find frontiers
n.trials	A string indicating the column in data which indicates the numer of trials for binomial routine where all element of n.trials > y
W.nb	A neighbourhood matrix indicating which elements of data are adjacent to each other. Default is set to NULL and extracted within the function unless overwritten
...	Other option to pass

Details

This is a wrapper other frontier detection functions. Return a list object of class frontier_model. A summary of the results can be called using summary(frontier_model). To extract the frontiers as sf line objects use frontier_as_sf(frontier_model).

Goal: standardise input and outputs regardless of technique? TODO: Check out how packages like matchit and synthpop deal with calling methods from other packages Orginally called inla_frontier in the project Returns list object of class: frontier_model

Value

Return a list object of class frontier_model and also INLA model.

W.frontiers An adjancency matrix where 1 is a border between units, 0 indicates a frontiers and NA are non-bordering areal units

data returns the data in the original call as a sf object with an id column

MengLeZhang/socialFrontiers documentation built on May 18, 2024, 9:36 a.m.