Brazil's Population

Creating a dataset about the Brazil's population from the last 20 yeas (1998 - 2017)
author: Rodrigo Silva

knitr::opts_chunk$set(echo = FALSE)

The goal of this project is just create a dataset about Brazil's population from the last 20 years to make easy any demographic processing. All the data was obtained from IBGE (Brazilian Institute of Geography and Statistics) and TCU (Federal General Accounting Office).

library(foreign)
library(dplyr)

setwd('Data/')

files <- list.files(pattern = '.DBF')

myfiles <- lapply(files, read.dbf)

populacao <- do.call(rbind, myfiles)

Nomenclature of municipalities and regions

municipios <- read.csv2('Municipios_Brasileiros.csv', header=TRUE, sep = ',')

regioes <- read.csv2('UF-Regiões.csv', header = TRUE)

Merging data about location

names(municipios)[3] <- 'CO_UF'

brasil <- merge(municipios, regioes, all.x = TRUE)

Filtering for the desired parameters

brasil <- brasil %>%
  select(Código.IBGE, Nome.do.Município, Estado, UF, NO_REGIAO, Latitude, Longitude)

Renaming columns

# Location
names(brasil)[1] <- 'code'
names(brasil)[2] <- 'city'
names(brasil)[3] <- 'state'
names(brasil)[4] <- 'abbreviation'
names(brasil)[5] <- 'region'
names(brasil)[6] <- 'lat'
names(brasil)[7] <- 'long'

# Population
names(populacao)[1] <- 'code'
names(populacao)[2] <- 'year'
names(populacao)[3] <- 'population'

The code about municipalities used to have 7 digits, the 7th digit was necessary to verify the code, so as we don't use it anymore I'll cut it out from the data.

brasil$code <- substr(as.character(brasil$code),1,6)

Other thing is bothering me is the name REGIAO in the region column, I gonna take it out too.

brasil$region <- substr(as.character(brasil$region), 8, 20)

Merging data about population with data about location

resultado <- merge(brasil, populacao, all.x = TRUE)

Spreading the result to better viewing

library(tidyr)

resultado_horizontal <- spread(resultado, year, population)

Exporting results

write.csv2(resultado, "brazil-population-2012-2017.csv", row.names = FALSE)

write.csv2(resultado_horizontal, "brazil-population-2012-2017-spread.csv", row.names = FALSE)

ASAP I'll continue improving this dataset! I really hope you enjoy this!