/DSTget

Primary LanguageR

DSTget

Get Statistics Denmark easily into R

DSTget makes it easy to download Statistics Denmark (DST) data straight into R and keep them updated. Its an interface to the flexible DST API.

How to install?

The package is not on CRAN, but can be installed directly from github.

library(devtools)
install_github("ogroendal/DSTget")

Features

DSTget has the following features:

  • Easily retrieve updated versions of tables without updating your code.
  • Set time periods as R dates
  • Get convenient extra variables - such as all statistical periods converted to R dates
  • Transform implicit DST datatypes into their R equivelants automatically - factors and dates are thus represented cleanly.
  • Make it easy to learn more about the contents of a table.
  • Get around download limits set in place by the standard web interface.
  • Makes it faster to construct table calls
  • Get more meaningful error messages

Get started DSTget assumes you know the name of the table you want.

Refer to statistikbanken.dk for an overview of the thousands of free and amazing tables on offer.

Downloading a table

If you just want the whole table, simply supply the table name. The below example is a long time series of divorces and marriages in Denmark.

The MyTableObject contains table metadata.

MyTableObject <- DSTget('BEV3C')
MyDataFrame <- getData(MyTableObject)

Exploring table metadata

You can check for yourself the metadata of a table

MyTableObject <- DSTget('BEV3C')
summary(MyTableObject) ## gives a convenient summary of the table

You can see all the possible values of a variable using the metadata object. First check the variables and then check the values

MyTableObject <- DSTget('HFUDD10')
head(MyTableObject$variables)
head(MyTableObject$values$HFUDD)

Specifying variables and values

If the table is large you can select only a few variables, cutting download times and complexity. The below example is the main population table, where we want the data summarized only on gender, marriage status and age. We subset the data on the variable CIVILSTAND, seeing only married or divorced individuals. We also specify that we want to see numbers from after the beginning of 2017. And that only for the ages of 10, 20 and 30 year olds.

MyTableObject <- DSTget('FOLK1A')
MyDataFrame <- getData(MyTableObject, CIVILSTAND = c("F", "G"),
 ALDER = c(10,20,30) ,  startDate = as.Date("2017-01-01"))

Fill out all remaining variables

If want almost all variables but dont want to specify them manually, then you can use the fillRemaining argument.

MyTableObject <- DSTget('FOLK1A')
MyDataFrame <- getData(MyTableObject, CIVILSTAND = c("F", "G"),
 ALDER = c(10,20,30) ,  startDate = as.Date("2017-01-01"), fillRemaining = T)

Now MyDataFrama also contains all the values for all variables not mentioned in the table specification.

Downloading large tables

Sometimes your table specification generates more than 100.000 rows. At which point the DST api will stop you. DSTget will conveniently split your table specification into a series of smaller downloads, and then give you one large table. Simplify specify the splitLarge argument. Be careful, downloading 250K rows or 5 million rows is totally fine for most computers and connections, but there are tables that are many many times bigger.

MyTableObject <- DSTget('FOLK1A')
MyDataFrame <- getData(MyTableObject, CIVILSTAND = c("F","G"), ALDER = 1:80,
 startDate = as.Date("2016-01-01"), fillRemaining = T, splitLarge = T)