Hitachi Vantara Pentaho Community Forums
Results 1 to 5 of 5

Thread: semi additive measure

  1. #1
    Join Date
    Jun 2012

    Default semi additive measure


    I have a measure that is additive across all dimensions other than time.Essentially, in the time dimension, it computes the number of open bookings, so
    what it should return for a given time slicing is the value of the booking at the Opening Period.
    As a calculated measure, something like (OpeningPeriod([Time].[Day]), [Measures].[Open Bookings]) returns the measure i want correctly and deals with roll ups as well., as long as queries are not filtered by time.

    For example, if the where clause has a date range, this calculation breaks down.

    Does anyone know how to correctly compute the opening value that respects filtering on time dimension?

  2. #2
    Join Date
    Mar 2007


    Mondrian doesn't support semi-additive measures for the time being. It is currently scheduled on our post 4.0 roadmap. You can track the issue here:
    Luc Boudreau
    aka. Luc le Magnifique
    aka. Monsieur Oui Oui

    Lead Engineer, Pentaho Corporation
    Twitter: luclemagnifique
    IRC: Monsieur_Oui_Oui@freenode

  3. #3
    Join Date
    Apr 2008


    Luc is correct, however there is a workaround. In one of our cubes we had IIF statements at each level that mattered to the calculation of our measure. Our example had to do with accounting system hiearchies and ratios over time. It is laborious, but you can do it. We created Level1, Level2, Level3, etc measures that would calculate accurately at each hiearchy level, so we could verify them. Then we had one VISIBLE measure that would use those IIF statements to test for the current level of the hierarchy being viewed, say "Year" or "Summary level 2" and would pick the Level2 measure.

    It's a bit nasty, but it works.

    Please, sign up for a JIRA account when you hit the link that Luc provided
    Then click on "Vote" to signify that it is an important feature to you. You enable the development team to prioritize tasks by popularity by voting for issues you care about.
    Last edited by Smoodo; 06-26-2012 at 03:27 PM.

  4. #4
    Join Date
    Jan 2008


    Hi Luc, this strategy is not working when using tools like JPIVOT or SAIKU. Once all of the dimensions are included in the mdx query, the measures keep appearing when you take one dimension away. I think this is because the level of a dimensions not included in the mdx query is 0, the same as level of a dimension like “Gender” included in the query. Because Pentaho does not have the AXIS function implemented, , you can not tell weather a dimension is included or not in a mdx query. Any suggestion ?

  5. #5
    Join Date
    Dec 2013

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
Privacy Policy | Legal Notices | Safe Harbor Privacy Policy

Copyright © 2005 - 2019 Hitachi Vantara Corporation. All Rights Reserved.