Skip to main content
Announcements
See why Qlik was named a Leader in the 2024 Gartner® Magic Quadrant™ for Data Integration Tools for the ninth year in a row: Get the report
cancel
Showing results for 
Search instead for 
Did you mean: 
thimiran1997
Contributor II
Contributor II

How Can I fix Partial sum error

There is  showing the partial sum according to part no but there is an error. How can I fix it. I attached the picture in below.

thimiran1997_0-1640667836156.png

 

#pivot #partial_Sum

7 Replies
tresesco
MVP
MVP

78 seems to be a negative here. Are you using any number formatting? May be the negative number is being displayed as positive too?

thimiran1997
Contributor II
Contributor II
Author

All are positive numbers

_Gerardo_
Partner - Contributor III
Partner - Contributor III

I always say "Qlik is nevers wrong".

This happens when there are duplicate values in the data, maybe the whole record of a table, I suggest you add the field "PART_NO" as a List Box and filter the values UA014008 and UA014009 with the option "Show Frequency" and then add all the fields that come from that table in a Table Box so you can compare the amount of different values that appear in the table versus the ones that the List Box counts.

Are you using  DISTINCT in some part of the formula? If you are using it, it is possible that for a single row the values are distinct but for group UA0140 within column 91-120 there are distinct values.

Could you please post your formula?

If you need more help you can write me and I will be glad to help you.

thimiran1997
Contributor II
Contributor II
Author

It seems there is no duplicate values . I will attach the qvw file. Please check it. month and year in order 10 and 2021 are use to selections.

_Gerardo_
Partner - Contributor III
Partner - Contributor III

Hello again, I found the difference.

2021-12-30_11-15-33.png

I created the field 61-90 mod with this formula (in green color)

Sum( {<Ageing ={"61-90"},TRANSACTION_CODE={'ARRIVAL'}>-<Ageing ={"61-90"},TRANSACTION_CODE={'RETCREDIT','RETWORK','INVSCRAP','SCPCREDIT'}>} Purchase_Qty)

You used that formula to check if the quantity is lower than Stock Qty (< Sum ( [Stock Qty]) . For testing purposes, I'm not considering the previous aging 0-30,41-60 because in both cases the values were 0.

As you can see in blue color, the condition "< Sum ( [Stock Qty])" is TRUE only for PART_NO UA014008 and UA014009 at row level ... but in the SUM for all the values grouped under StyleP UP UA0140 is FALSE, and Qlik is showing in the right place (TOTAL in red color) the value for group calculation.  

Depending what do you want as TOTAL value at Row and Column Level ... you could use Sum ( TOTAL <StyleP> [Stock Qty]) for the comparison ... or if you need something more advanced you could also try using Dimensionality() / SecondaryDimensionality() functions to use a different calculation for each Total Dimension Level.

Dimensionality() function shows the number of dimensions in a row ... and using this you can use a different calculation for TOTAL Dimension 1 for example AVERAGE, and TOTAL Dimension 0 for example a SUM.

SecondaryDimensionality() works similar for columns.

The difference in the calculation of the total is found in other columns as well, but it is not as evident as in the 91-120 column, for example, although the total appears in 61-90 as 1,494, the sum of the individual row values is 1,337.

_Gerardo_
Partner - Contributor III
Partner - Contributor III

Hello @thimiran1997 ,

Did the provided response help you?

thimiran1997
Contributor II
Contributor II
Author

Hello @_Gerardo_ ,

Sorry for the my late response. I was little bit busy. your clarification is confused me.  I want to sum of columns in styleS level. Could you give me a simple solution or example qvw file. 

Thank you for the help