Need help to resolve this problem

[#1]

SHOPID	CheckOut	EventTime	CHECKOUT with Max_EventTime	Count_CHK1	Count_CHK2
1	CHK1	5
1	CHK2	10	CHK2		1
2	CHK1	30	CHK1	1
2	CHK2	5
2	CHK2	10
3	CHK1	7
3	CHK2	9	CHK2		1
				1	2
				CHK1	CHK2

I came across a scenario as below and need your kind help to solve it in QlikView.

Scenario:

SHOPID is the number given to a 'Shopper' and the Checkout is the checkout number. EventTime is the time spent in each Checkout. Due to an error in location system, a shopper is seen in more than one checkout (i.e Shopper 1 in checkout 1&2 and shopper 2 in checkout 1 and twice in 2 etc.). The correct register is the one that shopper spent most of the time (i.e max EventTime per SHOPID).

Calculation:

Need to count the distinct SHOPID s per CheckOut and display it in a Bar Garph.

A spreadsheet is attached to make it easy to understand.

[Jason_Michaelides]

Hi,

If CheckoutWithMaxTime always has at least one value for each SHOPID then you could just use this as the dimension and check the "hide null values" box. Simple Count(DISTINCT SHOPID) then.

Hope this helps,

Jason

[#3]

Thanks Jason, I tried the following expressionas in the dimension but it didn't work.

=if(EventTime=aggr(max(EventTime),SHOPID),CheckOut)

Where am I making the mistake?

[jagan]

Hi,

Try this

=aggr(if(EventTime=max(EventTime), CheckOut),SHOPID)

Regards,

Jagan.

[#5]

Thanks Jagan, but it didn't work either.

[#6]

Attached a qvw with sample data.

[jagan · Accepted solution]

Hi,

Try this as dimension.

=aggr(FirstSortedValue(CheckOut, -EventTime),SHOPID)

Hope it works for you.

Regards,

Jagan.

[#8]

Hey Jagan, you're a genius, it worked.

Thank you so much.

[#9]

Dimension: SHOPID

Expression : =aggr(max(EventTime),SHOPID)