Please start any new threads on our new site at https://forums.sqlteam.com. We've got lots of great SQL Server experts to answer whatever question you can come up with.

All Forums  SQL Server 2000 Forums  SQL Server Development (2000)  MDX - distinct count problem

Author  Topic  frog314 Starting Member 6 Posts Posted - 2004-03-20 : 16:41:34 Hello! I have build in the cube Sales (FoodMart 2000 database) distinct count measure which count distinct clients (on id). When I use this measure to analysis how many clients buy some group of products not everything is okay. I try to answer for question: How many customers buy group of products? To make analysis I’m use Panorama program. Could YOU help me solve this problem? I thing I have to build some calculated member but I don’t know how to do that! Here is example (what’s wrong): DATABASE: FoodMart 2000 CUBE: Sales SELECTED DIMENSIONS: customers products MEASURE: distinct customers count EXAMPLE (Good Imported Beer) (USA customers) (distinct customers count) = 51 ---probably true (Good Light Beer) (USA customers) (distinct customers count) = 37 --- probably true (Good Imported Beer and Good Light Beer) (USA customers) (distinct customers count) = 88 FALSE! OR (All products) (USA customers) (distinct customers count) = 5581 = N.true ?? ?? (Drink products) (USA customers) (distinct customers count) = 3458 = N.true ?? ?? (Food products) (USA customers) (distinct customers count) = 5525 = N.true ?? ?? (non-cons. products) (USA customers) (distinct customers count) = 4468 = N.true ?? ?? But (Drink and Food products) (USA customers) (distinct customers count) = 9010 = false! (Drink and non-cons products) (USA customers) (distinct customers count) = 7953 = false! (Drink and non-cons food products) (USA customers) (distinct customers count) = 13478 = false! ************* Frog314 *************