I'm currently struggling with a problem that seems fairly simple on the surface. I suspect that the issue is more the way I'm thinking about the problem than the problem itself.
I have a set of Business Groups, a set of Applications and the intersection between the two:
| Bus. Group | | Application | | Bus Group | Application |
| A | | W | | A | X |
| B | | X | | A | Y |
| C | | Y | | B | Y |
| | Z | | B | Z |
| | | | C | W |
| | | | C | Y |
I've loaded all this into Qlikview and I have two list boxes: one showing the Business Groups and one showing the Applications. As you'd imagine, when I select a Buiness Group the list in the Application box shows me the associated Applications. If I choose two Business Groups e.g. A and C, I can see the Applications that are associated with either A or C (i.e. W, X and Y). What I'd like to be able to do is to see only the Applications that are associated with both A and C (I.e. Y).
My first thought on this is that I could load resident from the Bus Group and intersection tables to create copies in such a way that they join only to the Application table. I'd then create a second Business Group list box based on the copy of the Bus Group, allowing me to select Application A from the first list box and Application C from the second list box, which should restrict the contents of the Application list box to Y. However this approach is not very scalable - if I wanted to choose three Business Groups, I'd have to copy the tables again.
Is there a simpler (and neater) way of doing this? I have a feeling that set analysis may be involved but I'm yet to use this much.
Thanks for any help you can offer,
Damian
