Purpose
This query detects outliers in a dataset.
Pattern
This query calculates how far each value is from the mean in terms of standard deviation, then filtering for those that exceed a defined threshold.
Parameters
- TBL: Defines the name of the table containing the data.
- UKEY: Specifies the key of the table.
- VAL: Specifies the column with the values in question.
- MAXSTD: Defines the maximum standard deviation.
Usage
In this case we have a table with website clicks of 10 days. We apply the OUTL pattern with the value 2 for maxstd to get the days on which the clicks are more than 2 standard deviations away from the average.
Dataset
Parameter
- TBL = [DQ-OUTL.WebTraffic$]
- UKEY = Date
- VAL = Clicks
- MAXSTD = 2
Statement
Output
