Convert a data.table from year to time period.

convert_time_period returns a formatted data.table aggregating values from years into time periods. The required format of the input data is described in Details.

Usage

convert_time_period(
  .x,
  time_mapping,
  do_interp = FALSE,
  do_extrap = FALSE,
  do_past_extrap = FALSE,
  year_name = "year",
  value_name = "value",
  time_aggregate = "mean",
  na.rm = TRUE,
  verbose = FALSE
)

Arguments

.x

a well-formatted data.table.

time_mapping

a time mapping data.table.

do_interp

logical indicating whether linear interpolation should be done.

do_extrap

logical indicating whether constant extrapolation should be done.

do_past_extrap

logical indicating whether constant extrapolation should be done for past value.

year_name

string column name of year.

value_name

string column name of value.

time_aggregate

function to aggregate yearly values ("mean","sum","max").

na.rm

logical indicating whether missing values should be removed.

verbose

logical indicating whether running in verbose mode.

Value

a converted data.table

Details

The input data.table should contain the columns "year" and "value", while the other columns are are considered as id columns. The resulting data.table will have a column "t" instead of "year. Note that the period, year and value column names can be specified.

The time mapping between year and time period should be provided as a data.table with columns "year" and "t", and refyear for interpolation and extrapolation. In case of missing periods, these can be linearly interpolated or constantly interpolated. If several years are mapped into the same period, values are averaged or the function fun.aggregate is used.

See also

convert_table, convert_gdx.

Other conversion functions: convert_gdx(), convert_region(), convert_sqlite(), convert_table()

Examples

library(data.table)

# original data.table
dd <- data.table(year = 2005:2050, value = 1:46)

# Convert yearly time-serie into time period
convert_time_period(dd, "t30")
#>          t value
#>     <char> <num>
#>  1:      1     2
#>  2:      2     6
#>  3:      3    11
#>  4:      4    16
#>  5:      5    21
#>  6:      6    26
#>  7:      7    31
#>  8:      8    36
#>  9:      9    41
#> 10:     10    45