data-raw/cambodia_population/cambodia_population.R

In johnrbryant/bdefdata: Data from the book Bayesian Demographic Estimation and Forecasting

library(methods)
library(dplyr)
library(dembase)
library(readr)
library(tidyr)

province_levels <- c("Phnom Penh",
                     "Kandal",
                     "Kampot",
                     "Preah Sihanouk",
                     "Kep",
                     "Svay Rieng",
                     "Takeo",
                     "Kampong Cham",    
                     "Koh Kong",
                     "Prey Veng",
                     "Battambang",
                     "Pailin",    
                     "Banteay Meanchey",
                     "Kampong Speu",
                     "Kampong Chhnang",
                     "Siem Reap",
                     "Pursat",
                     "Kampong Thom",
                     "Mondul Kiri",
                     "Kratie",
                     "Otdar Meanchey",
                     "Ratanak Kiri",
                     "Stung Treng",
                     "Preah Vihear")


get_data <- function(name) {
    data <- read.csv(sprintf("data-raw/cambodia_population/%s.csv", name),
                     header = FALSE,
                     nrow = 551,
                     as.is = TRUE,
                     col.names = c("NULL", "age", "male", "female", rep("NULL", 5)),
                     colClasses = c("NULL", "character", "character", "character", rep("NULL", 5)))
    province <- data %>%
        filter(grepl("AREA", age)) %>%
        pull(male)
    ans <- data %>%
        filter(grepl("yrs|years", age)) %>%
        mutate(province = rep(province, each = 17)) %>%
        mutate(age = cleanAgeGroup(age),
               age = factor(age, levels = unique(age))) %>%
        gather(key = sex, value = count, -age, -province) %>%
        mutate(sex = factor(sex, levels = c("female", "male"), labels = c("Female", "Male"))) %>%
        mutate(count = as.integer(count))
    ans
}


cambodia_population <- bind_rows(get_data("urban"), get_data("rural")) %>%
    mutate(province = factor(province, levels = province_levels)) %>%
    xtabs(count ~ age + sex + province, data = .) %>%
    array(dim = dim(.), dimnames = dimnames(.))


save(cambodia_population,
     file = "data/cambodia_population.rda")