coin_win: Winsorisation helper function
In COINr: Composite Indicator Construction and Analysis

Description Usage Arguments Details Value See Also Examples

To be used inside treat() to avoid repetitions. Winsorises a numerical vector of data.

coin_win(
  icol,
  winmax,
  winchange = TRUE,
  t_skew = 2,
  t_kurt = 3.5,
  icode = NULL
)

`icol`	The vector of data to Winsorise
`winmax`	The maximum number of points to Winsorise for each indicator. If `NA`, will keep Winsorising until skewness and kurtosis thresholds achieved (but it is likely this will cause errors).
`winchange`	Logical: if `TRUE`, Winsorisation can change direction from one iteration to the next. Otherwise if `FALSE` (default), no change.
`t_skew`	Absolute skew threshold (default 2).
`t_kurt`	Kurtosis threshold (default 3.5).
`icode`	The indicator name - used for error messages in `treat()`.

Outliers are identified according to skewness and kurtosis thresholds. The algorithm attempts to reduce the absolute skew and kurtosis by successively Winsorising points up to a specified limit.

The process is detailed in the COINr online documentation.

A list containing:

.$icol the vector of treated data
.$imax the indices of elements of the vector that were Winsorised as high values
.$imax the indices of elements of the vector that were Winsorised as low values
.$winz the total number of Winsorised points

treat() Outlier treatment

# get a column of data with outliers
x <- ASEMIndData$Tariff
# Winsorise up to five points
winlist <- coin_win(x, winmax = 5)
# check the differences
data.frame(
Orig = x,
Treated = winlist$icol,
Changes = ifelse(x == winlist$icol, "Same", "Treated"))