MDX Essentials: Basic Numeric Functions: The Count() Function - Page 4

March 8, 2004

Excluding Empty Cells

The elimination of empty cells is a mere keyword away. We will accomplish our objective in the steps that follow:

'Count( {[Store Type].[Store Type].MEMBERS }, 
  EXCLUDEEMPTY)'

-- MDX17-2:  Counting that excludes Empty Cells

The Query pane appears as shown in Illustration 4.

The Results pane is populated, and the dataset shown in Illustration 5 appears. We see that the counts drop to two, five and five, for the Canada, Mexico and USA Store Types, respectively.

Let's tidy things up a bit by ordering our results dataset based upon the counts we have obtained.

    { [Store].[Store Country].MEMBERS } ON ROWS

{ORDER ( [Store].[Store Country].MEMBERS,
         [Store Type].[Types Count] , BDESC ) } ON ROWS

    -- MDX17-3:  Ordering the Row Axis

The Query pane appears as shown in Illustration 6.

The Results pane is populated, and the dataset shown in Illustration 7 appears. The results are ordered by Types Count, largest to smallest.

NOTE: For a tutorial surrounding the Order() function, see MDX Essentials: Basic Set Functions: The Order() Function.

We appear to have attained our objective. One more step, near and dear to the hearts of those who wish to ensure the quality of their work, and particularly to recovering CPA's like myself, is a test of accuracy and completeness. Let's look at the answer we are about to give Finance from another perspective - and at the same time look ahead to the subject of our next lesson, the CrossJoin() function.