Do not input private or sensitive data. View Qlik Privacy & Cookie Policy.
Skip to main content

Announcements
Qlik GA: Multivariate Time Series in Qlik Predict: Get Details
cancel
Showing results for 
Search instead for 
Did you mean: 
Not applicable

The abnormal data

Hi,

When I'm working with my qvw file, it occurs some abnomal data.

abnormal data.PNG.png

And I think my relevant code is all right, so I don't know where exists some issues.

Please help me! Thanks!

Best regards,

CoCo

1 Solution

Accepted Solutions
tresesco
MVP
MVP

Goto the definition of variable vOtherTotal, put '=' before the expression.

View solution in original post

9 Replies
tresesco
MVP
MVP

Qvd doesn't show data.  Could you share sample(reduced and scrambled) app?

Not applicable
Author

Related files.

Not applicable
Author

Related files.

Not applicable
Author

Related files.

Not applicable
Author

Hi tresesco,

I've attached the relevant files. Please hava a look. Thanks.

Best regards,

CoCo

tresesco
MVP
MVP

Goto the definition of variable vOtherTotal, put '=' before the expression.

Not applicable
Author

Oh, Great! Thanks too much.

But the variable "vMBTotal" is also like "vOtherTotal", and the corresponding result is right. So why?

normal data.PNG.png

tresesco
MVP
MVP

Because, in the vOtherTotal, you have two expression like Count(A)-Count(B), Ultimately if it gets expanded in the chart expression then becomes something like Exp1/Count(A)-Count(B), which is wrong, you actually need, Exp1/(count(a)-count(b)). but for vMBTotal, it is single count(), so there would not an issue if gets expanded in the chart, it would be Exp1/Count(a)... which is no different. If you put an equal sign in the variable definition, then the variable expansion would not cause the issue, because it is evaluated first. Hope it helps you understand.

Edit: for better understanding refer : http://community.qlik.com/blogs/qlikviewdesignblog/2013/11/04/the-magic-of-variables

Not applicable
Author

Hi tresesco,

You explain very good. I understand.  Thanks too much.

Best regards,

CoCo.