Reconstructing Improper Values with Local Centered Moving Trimmed Mean

Non-Causal Step

This step replaces any improper values in the data vector with the Trimmed Mean of the neighbourhood values and is resistant to outliers.

Input Parameters

1. Input data including improper values
2. Definition of improper values (‘nan’, ‘inf’, or ‘null’, etc.)
3. Percentage of high and low extremes to be trimmed
4. Local window size

Output Parameters

1. Output data excluding improper values

Workflow

Algorithm

Trimmed Mean

References

• J. Han, M. Kamber and J. Pei, Data Mining - Concepts and Techniques, 3rd ed., Amsterdam: Morgan Kaufmann Publishers, 2012.