This time we’re going bigger than ever. Fabric, Power BI, SQL, AI and more. We're covering it all. You won't want to miss it.
Learn moreDid you hear? There's a new SQL AI Developer certification (DP-800). Start preparing now and be one of the first to get certified. Register now
Hi
In our Power BI dashboard, we want a simple visualization as explained below.
The table consists of two columns: a date and the temprature that was measured on that date
A user has to select a time range, being a begin and end date. e.g. 01/01/2022 and 01/01/2023
Next we want to aggregate the temperatures within this range. We are talking about a custom aggregate function that is not one of the standard aggregation functions (like average, median). We want to do something more complex such as calculating the linear regression coefficients.
The function to calculate the linear regression coefficients has been implented by us in SQL and is registered in the underlying database. Eventually leading our query to be something like: SELECT linear_regression_coefs(temprature) in MyTable where date > start_date_param AND date end_date_param
Now, how am I able to implement this query and use my function in Power BI?
There are a few blog entries for that, for example Chris Webb's BI Blog: TSQL Table-Valued Functions And Dynamic M Parameters In Power BI DirectQuery (...
Check out the April 2026 Power BI update to learn about new features.
Sign up to receive a private message when registration opens and key events begin.
If you have recently started exploring Fabric, we'd love to hear how it's going. Your feedback can help with product improvements.