Trimmed Mean

Global Algorithm - One-Dimensional Algorithm

Trimmed Mean algorithm chops off values at the high and low extremes and applies the Mean algorithm to the remaining observations. In this way, the effect caused by a small number of extreme values can be offsetted.

Input Parameters

Parameter	Type	Constraint	Description	Remarks
$Y$	$Y \in \mathbb R^N$	$N \in \mathbb{N}$	Input data vector of length $N$
$p$	$p \in \mathbb R$	$0\% \lt p \lt 50\%$	Percentage of high and low extremes to be trimmed

Output Parameters

Parameter	Type	Constraint	Description	Remarks
$\mu$	$\mu \in \mathbb R$		Trimmed mean value of data $Y$	The result is sensitive to improper values such as ‘nan’, ‘inf’, ‘null’, etc.

Tool Support

• Excel

  For details refer to the online documentation of the function ‘TRIMMEAN’.

• Matlab

  For details refer to the online documentation of the function ‘trimmean’.

Single Steps using the Algorithm

• Computing Trimmed-Mean Value with Trimmed Mean Algorithm
• Outlier Detection with Trimmed Mean
• Reconstructing Improper Values with Local Centered Moving Trimmed Mean