Question
· Dec 14, 2015

Creating Trend line (scorecard) from data rather than counts

I need to create a KPI similar to the  DeepSee.Model.KPIs.DemoTrendLines kpi in Samples.  This kpi supports a scorecard widget with a trend line.  It shows patient counts by city with a trend line of the count of patients over a decade.  What I want is to show the trend line based on the average allergy score ([Measures].[Avg Test Score]).  A pivot table query that shows the data I would want to base the trend line on is:

SELECT NON EMPTY [BirthD].[H1].[Decade].Members ON 0,
NON EMPTY [HomeD].[H1].[City].Members ON 1 
FROM [Patients] %FILTER [Measures].[Avg Test Score]

Has anyone done this?

Discussion (3)2
Log in or sign up to continue

Found the solution.  The following MDX gives the values that I want.

SELECT {MEASURES.[Avg Test Score],
%LIST(NONEMPTYCROSSJOIN([BirthD].[H1].[Decade].Members,{[Measures].[Avg Test Score]}))} ON 0,
NON EMPTY homed.city.MEMBERS ON 1
FROM patients

I think the key was to enclose the measure in the NONEMPTYCROSSJOIN function in curly braces.  I had not done this in a previous attempt at getting this to work.