Calculate a Winsorized mean where extreme values are capped rather than discarded.
Winsorising preserves the sample size by capping extremes at the nearest retained value, unlike trimming which removes them. That matters when downstream variance calculations depend on n. Winsorising is preferred to trimming when the number of observations must be preserved, such as in panel data or repeated measures.
Winsorized Mean
Winsorized mean replaces the extreme p% with the nearest retained value, keeping n unchanged
Winsorized mean replaces the extreme p% with the nearest retained value, keeping n unchanged Winsorising preserves the sample size by capping extremes at the nearest retained value, unlike trimming which removes them. That matters when downstream variance calculations depend on n.
Winsorising is preferred to trimming when the number of observations must be preserved, such as in panel data or repeated measures.
This calculator takes 6 inputs: Sample size, Percentage Winsorized at each tail, Original mean, Value replacing the lower tail, Value replacing the upper tail, Sum of the untouched middle values. The pre-filled defaults are a realistic starting point — replace them with figures from your own environment for a result you can act on.