#055 How to Identify Data Outliers in SQL Server

We'll get to know the techniques:

1.

• A Common Table Expression (CTE) is a named temporary query that you can reference within a SELECT, INSERT, UPDATE, or DELETE statement. It makes it easy to write and read complex queries. Example:

  WITH CTEExample AS (
   SELECT Col1, Col2
   FROM YourTable
   WHERE SomeCondition
  )
  SELECT*
  FROM CTEExample;
  

2. Query Table Records (SELECT):

• SELECT is used to retrieve data from one or more tables in a database. Basic example:

  SELECT Column1, Column2
  FROM YourTable
  WHERE SomeCondition;
  

3. Calculate Quartile, Median, Interquartile, and Outliers (PERCENTILE_CONT):

• The PERCENTILE_CONT is used to calculate percentile values, useful for statistical analyses. Example:

  SELECT
   PERCENTILE_CONT(0.25) WITHIN GROUP (ORDER BY YourColumn) AS Quartile1,
   PERCENTILE_CONT(0.5) WITHIN GROUP (ORDER BY YourColumn) AS Median,
   PERCENTILE_CONT(0.75) WITHIN GROUP (ORDER BY YourColumn) AS Quartile3
  FROM YourTable;
  

4. Calculate upper and lower fences:

• Upper and lower fences are often used to identify outliers. They can be calculated using the quartiles obtained earlier.

configure in Power BI:

• In Power BI, go to the "Visualizations" tab and add the "Box plot" visual to your page. Configure it by selecting the appropriate columns for the axes.

(CROSS APPLY):

• CROSS APPLY is used to combine two tables, similar to INNER JOIN, but with some differences. Example:

  SELECT *
  FROM Table1
  CROSS APPLY Table2
  WHERE SomeCondition;
  

7. Conditional column (CASE, WHEN, ELSE):

• The CASE statement is used to create conditional columns. Example:

  SELECT
   Column1,
   CASE
      WHEN Column2 > 10 THEN 'High'
      WHEN Column2 > 5 THEN 'Medium'
      ELSE 'Low'
   END AS Priority
  FROM YourTable;
  

8. Identify the outliers of the records (Power BI):

• Use the Power BI graphical tools to visualize and identify the outliers in the Box Plot you added earlier. This will allow for easy visual analysis.