To address your requirements for data quality checks on the "Creation Date" field in the

item_branch__parquet_common_refined4

model using Elementary, here’s how you can approach each check: ### 1. Changes in the Average Date Age To monitor the "average date age" (i.e., the average difference between the current date and the "Creation Date"), you can use the

elementary.column_anomalies

test with the

average

monitor. For a date column, the "average" will be calculated as the average of the date values, which typically results in a timestamp or date. To get the "age," you would need to create a derived column (e.g.,

DATEDIFF(current_date, creation_date)

) and monitor its average. However, Elementary's built-in

average

monitor on a date column will just compute the average date, not the age in days. If you want to specifically monitor the age, you should create a new column in your model for this calculation and then apply the anomaly test to that column. ### 2. Number of Records with Dates Over Time You can use the

elementary.volume_anomalies

test with the

timestamp_column

set to "creation date." This will monitor the count of records per time bucket (e.g., daily) based on the "creation date" field, allowing you to detect spikes or drops in the number of records created over time. This is a recommended approach for tracking volume changes related to a date field [Volume anomalies test](https://docs.elementary-data.com/data-tests/anomaly-detection-tests/volume-anomalies). ### 3. Null Checks on "Creation Date" To check for nulls in the "creation date" field, you can use the

elementary.column_anomalies

test with the

null_count

or

null_percent

monitor. This will alert you if there is an unexpected increase in null values for this column [Column anomalies test](https://docs.elementary-data.com/data-tests/anomaly-detection-tests/column-anomalies). --- ### Example YAML Configuration

yaml models:   - name: item_branch__parquet_common_refined4     tests:       - elementary.volume_anomalies:           timestamp_column: creation_date           anomaly_direction: both           ignore_small_changes:             spike_failure_percent_threshold: 1             drop_failure_percent_threshold: 1           tags: ["elementary", "daily_elementary", "refined4", "common_refined4"]     columns:       - name: creation_date         tests:           - elementary.column_anomalies:               column_anomalies:                 - null_count                 - average       - name: "%itembranchskey"         quote: true         tests:           - unique:               config:                 severity: error                 error_if: ">1"                 warn_if: ">1"           - not_null

Note: - The

average

monitor on a date column will compute the average date, not the average age. For average age, create a derived column (e.g.,

age_in_days

) and monitor that. - Setting

timestamp_column: creation_date

in

volume_anomalies

will bucket your data by creation date, which is the recommended approach for time-based anomaly detection [timestamp_column documentation](https://docs.elementary-data.com/data-tests/anomaly-detection-configuration/timestamp-column). If you need to monitor the average age specifically, you should add a column to your model like:

sql SELECT   *,   DATEDIFF(current_date, creation_date) AS creation_date_age FROM ...

Then, add an anomaly test on

creation_date_age

with the

average

monitor. --- Summary: - Use

elementary.volume_anomalies

with

timestamp_column: creation_date

for volume over time. - Use

elementary.column_anomalies

with

null_count

for null checks. - Use

elementary.column_anomalies

with

average

for average date (or create an age column for average age). - For the average of a date column, Elementary computes the mean date, not the age; for age, use a derived column. References: [Column anomalies test](https://docs.elementary-data.com/data-tests/anomaly-detection-tests/column-anomalies) [Volume anomalies test](https://docs.elementary-data.com/data-tests/anomaly-detection-tests/volume-anomalies) [timestamp_column documentation](https://docs.elementary-data.com/data-tests/anomaly-detection-configuration/timestamp-column)