Anomaly Detection

See Anomaly Detection insights

Analytics regularly scans your data for anomalies, and surfaces those insights in your Dashboard.

If there are anomalies in any of your time-series data, you see hollow bubbles for those data points.

Hover over a bubble to see a summary of the anomaly.

 

Click a bubble to open an Insight card with detailed information in the right pane.

About Anomaly Detection

Anomaly Detection is a statistical technique to identify “outliers” in time-series data for a given dimension value or metric.

First, Analytics selects a period of historic data to train its forecasting model. For detection of daily anomalies, the training period is 90 days. For detection of weekly anomalies, the training period is 32 weeks.

Then, Analytics applies a Bayesian structural time series model to the historic data to forecast the value of the most recent observed datapoint in the time series.

Finally, Analytics flags the datapoint as an anomaly using a statistical significance test with p-value thresholds based on the amount of data in the reporting view.

Was this helpful?
How can we improve it?