Hacker News new | ask | show | jobs
by timeagain 978 days ago
But wouldn’t the trimmed mean mentioned in the article do this without “windsoriszing”?
1 comments

The trimmed mean discards outliers, so it can measure "what is the typical value for non-outliers?".

The windsorized mean reduces the weight of outliers, so it can measure "how many outliers are there?" instead of "how extreme are the outliers?"