Convert the time variable into either a date or numeric.

Time periods in data of CBS are coded: yyyyXXww (e.g. 2018JJ00, 2018MM10, 2018KW02), which contains year (yyyy), type (XX) and index (ww). cbs_add_date_column converts these codes into a Date() or numeric. In addition it adds a frequency column denoting the type of the column.

cbs_add_date_column(x, date_type = c("Date", "numeric"), ...)

Arguments

x

data.frame retrieved using cbs_get_data()

date_type

Type of date column: "Date", "numeric. Numeric creates a fractional number which signs the "middle" of the period. e.g. 2018JJ00 -> 2018.5 and 2018KW01 -> 2018.167. This is for the following reasons: otherwise 2018.0 could mean 2018, 2018 Q1 or 2018 Jan, and furthermore 2018.75 is a bit strange for 2018 Q4. If all codes in the dataset have frequency "Y" the numeric output will be integer.

...

future use.

Value

original dataset with two added columns: <period>_date and <period>_freq. This last column is a factor with levels: Y, Q and M

See also

Other data retrieval: cbs_add_label_columns(), cbs_add_unit_column(), cbs_download_data(), cbs_extract_table_id(), cbs_get_data(), cbs_get_data_from_link()

Other meta data: cbs_add_label_columns(), cbs_add_unit_column(), cbs_download_meta(), cbs_get_meta()

Examples

if (FALSE) { # \dontrun{
x <- cbs_get_data( id      = "7196ENG"      # table id
                 , Periods = "2000MM03"     # March 2000
                 , CPI     = "000000"       # Category code for total 
                 )

# add a Periods_Date column
x <- cbs_add_date_column(x)
x

# add a Periods_numeric column
x <- cbs_add_date_column(x, date_type = "numeric")
x
} # }