Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Hello!
I am currently working on one problem which seems to be complicated to me. I have monthly data which I need to compare. I always compare two neighbour month(9&10, 10&11,etc). Every month new items are added to the list of existing ones and I new to find and count them. My current table structure looks like this:
Table1. Total = 3 items
Item | Field1 | Field2 | MonthNo |
A | 1 | 1 | 10 |
B | 2 | 1 | 10 |
C | 2 | 3 | 10 |
Table2. Total = 6 items
Item | Field1 | Field2 | MonthNo |
A | 1 | 1 | 11 |
B | 2 | 1 | 11 |
C | 2 | 3 | 11 |
F | 1 | 2 | 11 |
D | 2 | 1 | 11 |
G | 2 | 2 | 11 |
So the task is to find new items added in each month (always comparing latest and previous months) which meet the criteria: Field1=2 and count them. Answer in this case(months 11 and 10): Items count=2 (Items D and G meet the criteria).
I can get the monthly total number of distinct items and also number of items where Field1=2 but I need to compare the actual items codes in each month, find new ones(that didn’t exist in previous month) and then count them. Is there any solution for this problem?
Try searching for new+lost: http://community.qlik.com/search.jspa?q=new+lost
You should encounter for example these discussions:
http://community.qlik.com/message/678525#678525
http://community.qlik.com/message/262719#262719
That said you can easily identify new items in the load script if your source data is already ordered by month:
LOAD
Item,
Field1,
Field2,
MonthNo,
if(not exists Item, 1, 0) as IsNew,
if(Field1 = Field2) as IsEqualField1Field2
FROM ...source.
You can count the new items with count({<IsNew={1},IsEqualField1Field2={1}>}distinct Item)