Service Analytics

8.Anomaly Detection

Z-value test for outlier detection

The Z-value Zi for the data point Xi is defined as follows
 
  • The Z-value test computes the number of standard deviations by which a data point is distant from the mean. This provides a good proxy for the outlier score of that point.
  • A good “rule-of-thumb” is to use Zi 3 as a proxy for the anomaly.
 
In the example, μ = 10 and σ = 2, so that points outside of [10-6,10+6] can be considered as anomalies.

Diskussion