Skip to main content
cancel
Showing results for 
Search instead for 
Did you mean: 
Not applicable

Setanalyis and AGRR does it effect without selections though Aggr is based on Dimension?

Hi

I asked in the other thread regarding this Problem but I made it easy here to make Understand. Can anyone Help me with what is going on here. Why If I select some dimension the Values are correct and if I dont select the values are wrong. AGGR() in my expression is responsible for that but How can I solve it??

See the attached example.

[View:http://community.qlik.com/cfs-file.ashx/__key/CommunityServer.Components.UserFiles/00.00.00.49.52/Forumproblem143.qvw]

Regards

Sravan

1 Solution

Accepted Solutions
Not applicable
Author

Hello,

take a look at my expressions for MinorStops and MinorStops%. I added some aggr()-functions. Secondly I replaced within one of the given aggr() the dimension ActivityDefinitionId with MachineDefinitionId because I assume that you want to aggregate all sum() in one expr. over the same dim.

Regards, Roland

View solution in original post

3 Replies
Not applicable
Author

Hello,

take a look at my expressions for MinorStops and MinorStops%. I added some aggr()-functions. Secondly I replaced within one of the given aggr() the dimension ActivityDefinitionId with MachineDefinitionId because I assume that you want to aggregate all sum() in one expr. over the same dim.

Regards, Roland

Not applicable
Author

HI roland,

Thanks for your answer. The trick is to add as many dimensions as needed. That helped. I still wonder why after adding Aggr(expression, Dim1) etc the MinorStop value for MachineDefinitionID OEE_DV_DEVNAME_45 is 0. I think Aggr does not take Negative Values. It worked with Aggr and without Dimensions but with Aggr and Dimension why it is getting the Previous correct value.

Vielen Dank
Regards

Sravan

Not applicable
Author

Hello Sravan,

yes, this is the trick (anyway one of my favorite functions) . I didn't check it right now, but as far as I know does aggr() ignore null-values.

So long, Roland