I'm aware of the gCentroid function from the rgeos package that gives the geographical centroid of a bunch of longitude, latitude points... but what if I wanted to assign a weight to those points, how would I find the weighted centroid?
e.g., Say I want to find the best place to meet up... I have 10 people in London, 2 in Leeds, 5 in Glasgow. How do I find the place that is the least combined effort?
You could just make a spatial object where each point represents a person and then have 10 points for London, 2 for Leeds and 5 for Glasgow. The gCentroid function will then weight appropriately.
Here's an example.
library(dplyr)
library(rgeos)
library(sp)
# Locations of cities
city_locations <- tibble(
Name = c("London", "Glasgow", "Leeds"),
lat = c(51.507222, 55.860916, 53.799722),
lon = c(-0.1275,-4.251433,-1.549167)
)
# More elegant way to repeat cities
people_counts <- c(rep('London', 10), rep('Glasgow', 5), rep('Leeds', 2))
# Code to join repeated cities with lat and lon
people <-
tibble(Name=people_counts) %>%
left_join(city_locations, by='Name')
# make a spatial points data frame
people_sp <- SpatialPointsDataFrame(people, coords = people[c('lon', 'lat')], proj4string = CRS('EPSG:4326'))
# find the centroid
pub <- gCentroid(people_sp)
pub$Name <- 'Centroid'
library(mapview)
# plot on a map
mapview(people_sp) + mapview(pub, col.regions='red')
Looks like somewhere just north of Derby.
