clickR (version 0.9.45)

fix_dates: Fix dates

Description

Fixes dates. Dates can be recorded in numerous formats depending on the country, the traditions and the field of knowledge. fix.dates tries to detect all possible date formats and transforms all of them in the ISO standard favored by R (yyyy-mm-dd).

Usage

fix_dates(
  x,
  max.NA = 0.8,
  min.obs = nrow(x) * 0.05,
  use.probs = TRUE,
  select = 1:ncol(x),
  track = TRUE,
  parallel = TRUE
)

Arguments

x

A data.frame

max.NA

Maximum allowed proportion of NA values created by coercion. If the coercion to date creates more NA values than those specified in max.NA, then all changes will be reverted and the variable will remain unchanged.

min.obs

Minimum number of non-NA observations allowed per variable. If the variable has fewer non-NA observations, then it will be ignored by fix.dates.

use.probs

When there are multiple date formats in the same column, there can be ambiguities. For example, 04-06-2015 can be interpreted as 2015-06-04 or as 2015-04-06. If use.probs=TRUE, ambiguities will be solved by assigning to the most frequent date format in the column.

select

Numeric vector with the positions (all by default) to be affected by the function

track

Track changes?

parallel

Should the computations be performed in parallel? Set up strategy first with future::plan()

Examples

mydata<-data.frame(Dates1=c("25/06/1983", "25-08/2014", "2001/11/01", "2008-10-01"),
                   Dates2=c("01/01/85", "04/04/1982", "07/12-2016", "September 24, 2020"),
                   Numeric1=rnorm(4))
fix_dates(mydata)