In ropengov/sorvi: Functions for Finnish Open Data

Maailmanpankki tarjoaa runsaasti avointa dataa. Alla muutamia esimerkkeja aineistojen hakuun ja kasittelyyn R-kielella. Lisaa esimerkkeja taalta ja taalta.

Suomalaisten muuttoliiketilastot

Muuttotiedot maittain vuonna 2010, julkaistu Maailmanpankin sivuilla.

# sorvi installation: http://louhos.github.com/sorvi/asennus.html
library(sorvi)

GetWorldbankMigration <- function(countries) {

  library("gdata")

  # Load migration matrix from World Bank
  tmp <- try(migration.matrix <- gdata::read.xls("http://siteresources.worldbank.org/INTPROSPECTS/Resources/334934-1110315015165/T1.Estimates_of_Migrant_Stocks_2010.xls"))

  if (class(tmp) == "try-error") {stop("gdata::read.xls requires installation of the PERL module for Windows")}

  data.inds <- 2:214

  # Get migration flow in to and out from the given countries
  migration.dat <- list()
  for (i in 1:length(countries)) {
    message("Loading migration data for:", countries[i])
    if (!any(migration.matrix[1,]==countries[i])) {
      cat("- Error: country not found!\n")
    } else {
      migration.in <- as.vector(migration.matrix[data.inds, migration.matrix[1,]==countries[i]])
      migration.in <- as.numeric(gsub(",", "", migration.in))
      migration.out <- as.vector(as.matrix(migration.matrix[migration.matrix[,1]==countries[i], data.inds]))
      migration.out <- as.numeric(gsub(",", "", migration.out))
      migration.dat[[i]] <- data.frame(In=migration.in, Out=migration.out)
      message("- DONE\n")
    }
  }
  if (length(migration.dat) > 0)
    names(migration.dat) <- countries
  else
    stop("No data found - quitting\n")

  # Add country name information
  migration.dat$Country <- as.vector(migration.matrix[data.inds, 1])

  # Add manually alternative country names for some countries (based on names used in rworldmap)
  countries.alternative <- as.vector(migration.matrix[data.inds, 1])
  alt.names <- c("Bahamas", "Channel Islands", "Democratic Republic of the Congo", "Congo",
                 "Cote d'Ivoire", "Egypt", "Faroe Islands", "Gambia",
                 "Hong Kong", "Iran (Islamic Republic of)", "Korea, Democratic People's Republic of", "Korea, Republic of",
                 "Kosovo", "Kyrgyzstan", "Lao People's Democratic Republic", "Libyan Arab Jamahiriya",
                 "Macau", "The former Yugoslav Republic of Macedonia", 
                 "Micronesia, Federated States of", "Republic of Moldova",
                 "Burma", "Russia", "Sao Tome and Principe", "Slovakia",
                 "Saint Kitts and Nevis", "Saint Lucia", "Saint Vincent and the Grenadines", "United Republic of Tanzania",
                 "Venezuela", "Viet Nam", "United States Virgin Islands", "Palestine",
                 "Yemen")
  countries.alternative[c(14, 39, 44, 45, 47, 57, 63, 69, 84, 89, 101, 102, 103, 
                          105, 106, 111, 115, 116, 128, 129, 135, 160, 164, 171, 
                          178, 179, 180, 188, 207, 208, 209, 210, 211)] <- alt.names

  migration.dat$CountryAlternative=countries.alternative

  return(migration.dat)

}


migration <- GetWorldbankMigration("Finland")

Suomalaisten nettomuuton visualisointi maailmankartalla

Suomalaisten nettomuuton visualisointi maailmankartalla Louhos-blogista:

# Load worldmap
#install.packages("rworldmap")
library(rworldmap)
worldmap <- getMap(resolution="medium")

# Compute the total amount of migration in to and out from Finland
fin.in <- migration$Finland$In
fin.out <- migration$Finland$Out
fin.in[is.na(fin.in)] <- 0
fin.out[is.na(fin.out)] <- 0
migration.total <- rowSums(cbind(fin.in, fin.out), na.rm=T)

# Compute ratio indicating the direction of the migration 
# in to (1) or out from (0) Finland
migration.ratio <- (migration.total - fin.out)/migration.total
migration.ratio[migration.ratio==Inf] <- 1

# Total amount of migration is mapped to the strength of the colours
# Use logarithmic scale to reduce the effect of very large numbers (Sweden)
migration.total.log <- log(migration.total)
migration.total.log[migration.total.log==-Inf] <- 0
alpha <- migration.total.log/max(migration.total.log)

# Direction of migration is mapped to red and blue colours
in.ratio <- migration.ratio
out.ratio <- 1 - migration.ratio
in.ratio[is.nan(migration.ratio)] <- out.ratio[is.nan(migration.ratio)] <- 0

# Construct a rgb color scheme based on these values
cols.rgb <- rgb(red=in.ratio, green=0, blue=out.ratio, alpha=alpha)
cols.rgb[cols.rgb=="#00000000"] <- "grey90"
names(cols.rgb) <- migration$CountryAlternative

# Initialize colours for all countries in the worldmap to light grey
cols.countries <- rep("grey90", length(levels(worldmap@data$NAME)))
names(cols.countries) <- levels(worldmap@data$NAME)

# Map countries in the migration data set to the worldmap and update 
# country colours
mapping <- match(names(cols.rgb), names(cols.countries))
cols.countries[mapping[!is.na(mapping)]] <- cols.rgb[!is.na(mapping)]
cols.countries["Finland"] <- "black"

# Plot the map with the final visualization
q <- spplot(worldmap, "NAME", col.regions = cols.countries,
main = NULL, colorkey = FALSE, lwd = .4, col = "black")
print(q)

Suomen CO2-paastot

Seuraava esimerkki hakee Suomen CO2-paastotasot vuosilta 1990-2008 (Lahde: Louhos-blogi):

# Lataa WDI-paketti
#install.packages("WDI")
library(WDI)

# Poimi tutkittava indikaattori manuaalisesti
k <- 27
indi <- WDIsearch()[k,1]
titl <- WDIsearch()[k,2]

# Hae tiedot
df <- WDI(country="all", indicator=indi, start=1990, end=2008)

# Visualisoi
library(ggplot2)
theme_set(theme_bw(20)) 
p <- ggplot(data = subset(df, country == "Finland")) + aes(x = year, y = EN.ATM.CO2E.KD.GD) + geom_line() + opts(title = titl)
print(p)