To set up a test that finds volume anomalies for each day using the column

transaction_date

and applies day-of-week seasonality, you can use the following configuration in your dbt model:

yaml models:   - name: <your_model_name>     tests:       - elementary.volume_anomalies:           timestamp_column: transaction_date           seasonality: day_of_week           time_bucket:             period: day             count: 1

This configuration will compare each day's volume to previous days with the same day of the week (e.g., all Mondays compared to previous Mondays), which helps account for weekly patterns in your data. The

timestamp_column

should be set to

transaction_date

to group the data by day, and

seasonality: day_of_week

ensures the comparison is made against the same weekday in prior periods. The

time_bucket

is set to daily, which is required for day-of-week seasonality to work as intended. For more details, see the official documentation on [seasonality configuration](https://docs.elementary-data.com/data-tests/anomaly-detection-configuration/seasonality) and [volume anomalies test configuration](https://docs.elementary-data.com/data-tests/anomaly-detection-tests/volume-anomalies#test-configuration).

@kapa.ai Is it possible to make a day of month seasonality too?

I'm sorry, but the knowledge sources do not mention support for "day of month" seasonality in Elementary's anomaly detection tests. The only seasonality options explicitly documented are

day_of_week

,

hour_of_day

, and

hour_of_week

 [seasonality configuration](https://docs.elementary-data.com/data-tests/anomaly-detection-configuration/seasonality). There is no reference to a "day of month" seasonality parameter or example in the available documentation or community discussions.