Filter visual using dynamic aggregate calculation that references other slicer defined values

[Rob92]

Our dataset looks at loans.  I have a matrix that looks at average interest rate per county.  I currently have three slicers working on it - Date, Interest Rate, Loan Amount.  These work on the row level so there is no issue there.  My problem is that we need the ability to filter out counties with an average of less than N loans per month in the selected Date, Interest Rate & Loan Amount range.  Ideally the end user would have the ability to determin in a slicer the minimum loan-per-month low threshold for a county to be included, but at the very least I need the ability to pre-program it to exclude counties with less than say 10 loans per month in the selected range.  

If this was Excel, I could achieve this with a column on the source data table (using lookup/countifs/if) which counts rows per county where the date is in the selected date range (and meets other criteria) divided by months in the range, and shows 'Exclude' if that value is less than the loan-per-month threshold the user has defined, 'Include' if not; then filter the visual on that, and add a macro to refresh the data when the user changes the loan-per-month threshold.

Any ideas how I can do this in PowerBI?  The incompatability of aggragate-based measures and row-based column-calcs is the real issue here.  I need a column calculation that references an aggregate measure which reacts to user defined variables.  After a lot of research and testing, I'm totally stumped trying to provide what seems like pretty basic functionality to my clients.  Any help would be greatly appreciated! 

[Rob92 · Accepted solution]

I think I've found my solution!!  I'm yet to implement it, but in principle I can't see why it won't work:  Integrate a Power App into the Power BI report that writes user input back to the data source and updates with Direct Query in real time:  https://www.youtube.com/watch?v=us0hPFwQ4Zs

Will report back with anything noteable once I've put this into practice.

[lbendlin]

Note that this will require all your report users to have a PowerApps Premium license.

[lbendlin]

You cannot influence calculated columns from measures. That is a fundamental design decision/limitation in Power BI.  You may need to go back to using Excel.

[Rob92]

Thanks for your response.  I am accutely aware of that but it is not an option to go back to excel due to a number of other mission-critical reasons.  So I am trying to think outside the box here (and asking the community for help doing so), to find a way of achieving my desired result (first paragraph of my question) that bypasses Power BI's inability influence calculated columns from measures.