Winsorizing a vector means that a predefined quantum of the smallest and/or the largest values are replaced by less extreme values. Thereby the substitute values are the most extreme retained values.
winsorize(x, thresh = c(0.05, 0.95), na.rm = FALSE)
Numeric vector. Distribution to be winsorized.
Numeric vector. Lower and upper quantiles thresholds. Set values within [0,1].
Boolean. Should NA
be omitted to calculate the quantiles?
Note that NA
in x
are preserved and left unchanged anyway.
Numeric vector transformed.
Other Outliers:
outlier_turkey()
,
outlier_zscore_plot()
,
outlier_zscore()