Earn the coveted Fabric Analytics Engineer certification. 100% off your exam for a limited time only!
Hi!
I've been trying to do what said in this post http://community.powerbi.com/t5/Community-Blog/DAX-as-a-Language/ba-p/22120 with my own data, the problem is that I did all process and I have the correlation but I need that if I do some filter the correlation has to be re calculate, it's because for example I need to find the correlation in different cities. So I changed your code:
//First X CorrelateX3 = VAR AverageX3= AVERAGEX(ALLSELECTED('Llamadas Une'),[Valor]) RETURN ([Valor]-AverageX3) //Then Y CorrelateY3 = VAR AverageY3 = AVERAGEX(ALLSELECTED('TRPS Une Hogares'),[Trps P Mas 25 MA]) RETURN ([Trps P Mas 25 MA]- AverageY3) //Finally the corelation Correlation2 = VAR SumCorrelateX3SumCorrelateX3= SUMX(ALLSELECTED('Llamadas Une'),[CorrelateX3]*[CorrelateX3]) VAR SumCorrelateX3SumCorrelateY3= SUMX(ALLSELECTED('Llamadas Une'),[CorrelateX3]*RELATED('TRPS Une Hogares'[CorrelateY3])) VAR SumCorrelateY3CorrelateY3= SUMX(ALLSELECTED('Llamadas Une'),RELATED('TRPS Une Hogares'[CorrelateY3])*RELATED('TRPS Une Hogares'[CorrelateY3])) RETURN ( SumCorrelateX3SumCorrelateY3 / SQRT(SumCorrelateX3SumCorrelateX3*SumCorrelateY3CorrelateY3))
I calculated the x average to testing:
//no filter AverageX = AVERAGEX(ALL('Llamadas Une'),[Valor]) //Filter, it is like AverageX3 Average lla = AVERAGEX(ALLSELECTED('Llamadas Une'),[Valor])
I found that the error is because the subtraction between [Valor]-AverageX3 it's not ok:
As you can see here, the result in CorrelateX3 is equal to subtract Valor-AverageX and not Valor-AverageX3. I don't know why it's happend here, Colud you Help me with this issue? or do you have other idea to do a dinamic correlation?
Thanks
User | Count |
---|---|
139 | |
113 | |
103 | |
73 | |
63 |
User | Count |
---|---|
136 | |
125 | |
107 | |
70 | |
61 |