User Profile
rh203
Helper I
Joined 6 years ago
User Widgets
Contributions
Percentile, ignoring date dimension for graph
Hi, I have a data set for which I cam trying to calculate a percentile of the total values for a time horizon with the intention of then positoning it against the data in a chart to see which periods go above or below the quartile (percentile) for the full period of data. Unfortunately the measure appears to change along each of the dates instead of remain constant... I assume this is something to do with it factoring in the date dimension table... Data model (redacted - sorry - please let me know if any fields are needed) and measure below... Has anyone any ideas? black bar isRe: SSAS import - very strange behaviour!
Hi, sorry but you're incorrect - if you choose to connect using live connection (directquery) this is true, but you can also connect to an SSAS (same as you can to an SQL) using import mode to bring the data into the PBI model to them combine with other data elements...SSAS import - very strange behaviour!
Hi all, In the interest of time (I'll see first from the community what details are required to help out!) I won't get into a huge amount of details on the query and source data other than it is an SSAS source with multiple tables (one FACT table) and multiple DIM tables. I am trying to use power query to import data from the source in order to allow me to combine and customise in various reports I need to build for my org. I am using a DAX query (as I have done many times before) to drive the import criteria from the SSAS via power query. I have created a basic view of the data that I want to import within a blank PBIX file (simply as a sandbox to visualise the filters and expected data output and to use as a comparison for the data when it is extracted from the SSAS with my DAX query). The DAX statement consists of a basic SUMMARIZECOLUMNS with 'group by', 'filter' and 'summarize' calculations precisely matching the fields and criteria I have applied in the 'sandbox' PBIX but am completely baffled that the results don't remotely match the PBIX when the extract is complete. I have tried multiple tweaks to the DAX statement and nothing makes it any closer to the PBIX. The even more bizarre thing is that I have tried to import the data without a DAX query (using the cube navigation / folding functionality in Power Query) but still the data doesnt match. Is there something I am missing (I'm not familiar with SSAS backend so not sure if there's some magic happening there that I can't see but doesn't materialise using DAX import functionality). Appreciate this may be a vague without data but looking for some top level principles or pointers before spelling out the data structure and examples further.. Thanks all!Re: Parameter with Query as Source
Hi Jimmy, Everything in power query (excel to be precise). The 'variable' (single value) is of type whole number. I tried what you proposed and get the below messagw - think I understand the reason but not clear on how to pass the value into the query as a number instead? ThanksRe: Parameter with Query as Source
Thanks Jimmy, Still a bit stuck.. I am using the below filter statement within a DAX query to send to the SSAS to retreive the data required but when I run it, I get the error message below also. I'm using a single value (instead of a list) so not sure if I have used the correct DAX to drive this? FILTER('PFR Date','PFR Date'[PFR Week]={SELECTED_PFR_WEEK}),
