Negect current seletion in a complicated set analysis expression

[#1]

Hi, I have the following piece of data and what i would like to achieve is illustrated also.

Customer_no	Bank No	Transaction Date	Transaction Type	Amount
C0001	B0001	1/1/2015	A	1000
C0001	B0001	1/1/2015	B	2000
C0001	B0001	1/1/2015	C	2000
C0002	B0002	1/1/2015	A	3000
C0002	B0002	1/1/2015	B	3000
C0002	B0002	1/2/2015	A	4000
C0002	B0003	1/3/2015	A	4000
C0002	B0004	1/3/2015	B	4000
C0003	B0005	1/1/2015	A	5000
C0003	B0005	1/1/2015	B	5000
C0004	B0006	1/1/2015	A	6000
C0004	B0006	1/1/2015	A	6000
C0004	B0006	1/1/2015	B	8000
C0004	B0007	1/1/2015	A	8000
C0004	B0007	1/1/2015	B	1000
C0004	B0008	1/1/2015	A	1000
C0004	B0008	1/1/2015	B	8000
C0004	B0008	1/1/2015	A	8000
C0004	B0009	1/1/2015	A	1000
C0004	B0009	1/1/2015	B	8000
C0004	B0009	1/1/2015	C	8000

Basically, user can input a number in the input box to indicate the maximum number of transaction a bank account number could have per day, and the scattered plot will display the customer number with the bank account that hit the criteria.

My problem now is that I want to keep the graph the same when each dot is selected. Usually this can be achieved by disregarding the dimension in set analysis. In this case this will be {<Customer_no=>}.

However, it doesn't work this time, This is expression bank_no_transation_date={"=aggr(count({<Customer_no=>}distinct [Transaction Type]),bank_no_transation_date)>=$(vLimit)"}

I have attached the qvw file. ould you please advise?

[tresB]

Check attached qvw. Remove the Aggr() from your set expression section.

[#3]

Thanks. it works. I would also like to know in case aggr unction must be used, is there any alternative way of doing it apart from removing the aggr function?

[tresB]

Yes, you can use only() along with aggr() like:

bank_no_transation_date={"=Only({<Customer_no=>} aggr(count({<Customer_no=>}distinct [Transaction Type]),bank_no_transation_date))>=$(vLimit)"}

But, here too, I don't see the obviousness of using aggr().

I am really yet to find a case when aggr() usage is really needed in Set Analysis.