Forum Discussion

Hammarberg's avatar
Hammarberg
Frequent Visitor
9 years ago

How to calculate Forecast accuracy

                      Hi,   I've been trying to get my head around how to calculate a forecast accuracy. Here is how I usually set up a simply formula in excel:   First I calcula...
Need Help
Sean's avatar
Sean
Icon for Community Champion rankCommunity Champion
9 years ago

Hammarberg

If you want a MEASURE instead of COLUMN use this the SUMX function is an iterator

ABS Variance MEASURE = SUMX('Table', ABS('Table'[Forecast] - 'Table'[Sold]) )

Accuracy MEASURE = 1 - ( [ABS Variance MEASURE] / SUM('Table'[Forecast] ) )

This works with the first example you posted! (You can filter the 'Table' in the SUMX if necessary...)

To get anyone to help you further if this Measure doesn't help you with you actual data

Post sample data that can be copied and pasted into PBI not images!

Otherwise you are asking people to spend the time to recreate your data in order to test it and many will skip over your question

Hammarberg's avatar
Hammarberg
Frequent Visitor
9 years ago

Hi Sean,

 

The solution you had for ONE table works perfectly. I've tried some different approaches but hit a dead end. Attached you find a bigger excel sample file and a pbix. If you would be ready to have a look it would be greatly appreciated.

 

Example.xlsx

Sample.pbix

 

Let me know if there is any issues with the links (did not see how to upload files directly to the message).

  • hymieho's avatar
    hymieho
    Icon for Resolver I rankResolver I
    9 years ago

    When all values are in one table you can use the measures below. I couldn't figure out how to do it when each piece is in a different table. I tried SUMMARIZECOLUMNS but I couldn't get it to work...my attempt is at the bottom, maybe someone else can advise.

     

    Measures for Same Table

    Note: there was no "article" or 'date' column, and the join to 'Data in same model 2' confuses me. If you were to add Date and Article, you would probably want to summarize by those columns first.

     

    Accuracy New =
    VAR ABSErr =
    SUMX(
    'Data in same table 1',
    CALCULATE (
    ABS (
    SUM ('Data in same table 1'[Forecast]) - SUM ('Data in same table 1'[Sold])
    )
    )
    )
    VAR Volume = SUM('Data in same table 1'[Sold])
    VAR Accuracy = 1-(DIVIDE(AbsErr,Volume))

    RETURN
    IF(Volume=0,0,
    IF(Accuracy<=0,0,
    Accuracy)
    )

     

    Bias New =
    VAR Volume = SUM('Data in same table 1'[Sold])
    VAR Forecastvar = SUM('Data in same table 1'[Forecast])
    RETURN
    IF(Volume=0,IF(Forecastvar=0,0,1),(SUMX('Data in same table 1','Data in same table 1'[Forecast] - 'Data in same table 1'[Sold]) / Volume))

     

     

    Measure for values in different tables Not Working

    Forecast Accuracy =
    VAR ABSErr =
    SUMX(
    SUMMARIZECOLUMNS(
    'Dates'[Date],
    'Articles'[Article]),
    ABS (
    SUM (Forecast[Forecast]) - SUM (Sales[Sold])
    )

    )

    VAR Volume = SUM(Sales[Sold])
    VAR Accuracy = 1-(DIVIDE(AbsErr,Volume))

    RETURN
    IF(Volume=0,0,
    IF(Accuracy<=0,0,
    Accuracy)
    )