Hampel Identifier

Global Algorithm - One-Dimensional Algorithm

Hampel Identifier algorithm decides whether a value is a value outside the region of interest based on that value’s distance from the estimated median distribution. Generally, if

$$|Y_i - \text{Median}(Y)| > 3 \cdot \text{MAD} \, \text{,}$$

the data point is considered outside the region of interest. $\text{Median}(Y)$ is the Median, $Y_i$ is $i$th element of $Y$ and $\text{MAD}$ corresponds to the Median Absolute Deviation value.

Input Parameters

Parameter	Type	Constraint	Description	Remarks
$Y$	$Y \in \mathbb R^N$	$N \in \mathbb{N}$	Input data sequence of length $N$

Output Parameters

Parameter	Type	Constraint	Description	Remarks
$\hat{Y}$	$\hat{Y} \in \mathbb R^N$		Values in the $Y$ list which are outside the region of interest are marked

Tool Support

• R

Single Steps using the Algorithm

• Outlier Detection with Hampel Identifier

References

• R.K. Pearson, Mining Imperfect Data, Society for Industrial and Applied Mathematics, Philadelphia, PA., 2005.