Skip to main content
cancel
Showing results for 
Search instead for 
Did you mean: 

Earn the coveted Fabric Analytics Engineer certification. 100% off your exam for a limited time only!

Reply
Buidoistreet
Frequent Visitor

Get MEDIAN for first 10 months of three years of data

Hello,

 

My requirement is to pull three years of data from today, the ask is to display length of stay over this three years of data by month (YYYYMM) in a line chart.  In addition, add MEDIAN length of stay as a dash line, the calculation for MEDIAN is only take median of the first 10 starting months of data as circled on 2nd table screenshot.

 

median line.jpg
 

los.jpg

To do this, I created a RANK column below as shown on the above table:

Rank = rankx(ALLSELECTED(eras), ERAS[DISCHARGE_YYYYMM],,DESC,dense)

And add LOS_MEDIAN_FIRST_10PTS measure below as shown on the above table:
LOS_MEDIAN_10PTS = CALCULATE(MEDIANX(ERAS, AVERAGE(ERAS[LOS_EST_DAYS])), ERAS[Rank]>=27 && ERAS[Rank] <=36 , ALL(ERAS[DISCHARGE_YYYYMM]))

This appears to be working fine for initial requirement. However, the new requirement comes in and this is how the problem starts.
If I add a DATE slicer (users can choose any dates), my rank column doesn't start from rank = 1.
It displays oringal rank, not dynamically. So my LOS_MEDIAN_10PTS measure logic above will not work.
 
date slicer.jpg
I tried to add RANK measure below, and this measure ranks correctly starting from rank 1 and count up.
    RANK_TEST = RANKX ( ALLSELECTED ( ERAS[DISCHARGE_YYYYMM] ),
    CALCULATE ( SELECTEDVALUE ( ERAS[DISCHARGE_YYYYMM] ) ),
    ,
    ASC, Dense
)
 
However, I couldn't add this measure to replace a rank filter in red below.

LOS_MEDIAN_10PTS = CALCULATE(MEDIANX(ERAS, AVERAGE(ERAS[LOS_EST_DAYS])), [RANK2]>=1 && [RANK2]<=10 , ALL(ERAS[DISCHARGE_YYYYMM]))
 
Please advise. Thanks for your time and help!
 

 

1 ACCEPTED SOLUTION
v-frfei-msft
Community Support
Community Support

Hi @Buidoistreet ,

 

Please create a measure as below to work on it instead of using a calculated column.

Measure = 
VAR mindate =
    CALCULATE ( MIN ( 'ERAS'[date] ), ALLSELECTED ( ERAS ) )
VAR maxdate =
    EDATE ( mindate, 10 )
RETURN
    CALCULATE (
        AVERAGE ( ERAS[Length_of_Stay] ),
        FILTER ( ALL ( ERAS ), 'ERAS'[date] >= mindate && 'ERAS'[date] < maxdate )
    )

Capture.PNG

 

For more details, please check the pbix as attached.

 

Community Support Team _ Frank
If this post helps, then please consider Accept it as the solution to help the others find it more quickly.

View solution in original post

6 REPLIES 6
v-frfei-msft
Community Support
Community Support

Hi @Buidoistreet ,

 

Please create a measure as below to work on it instead of using a calculated column.

Measure = 
VAR mindate =
    CALCULATE ( MIN ( 'ERAS'[date] ), ALLSELECTED ( ERAS ) )
VAR maxdate =
    EDATE ( mindate, 10 )
RETURN
    CALCULATE (
        AVERAGE ( ERAS[Length_of_Stay] ),
        FILTER ( ALL ( ERAS ), 'ERAS'[date] >= mindate && 'ERAS'[date] < maxdate )
    )

Capture.PNG

 

For more details, please check the pbix as attached.

 

Community Support Team _ Frank
If this post helps, then please consider Accept it as the solution to help the others find it more quickly.

That helps. Thanks so much!

amitchandak
Super User
Super User

See if this way Rank can help


new Rank =
var _min = mix(allselected(ERAS),[Rank])
return
[Rank]-_min +1


Thanks atmitchandak for a quick reply!

 

I created a new rank as column and I've received this error "The expression refers to mutiple columns.  Multiple columns cannot be converted to scalar value."

 

new Rank =

var _min = min(allselected(ERAS), ERAS[Rank])

return

ERAS[Rank]-_min +1

 

Please advice.  Thanks!

Oh, there was a typo in MINX, it should be

 

new Rank =

var _min = minx(allselected(ERAS), ERAS[Rank])

return

ERAS[Rank]-_min +1

 

However, the new rank value still look same as the old rank. It doens't dynamically re-rank.

Greg_Deckler
Super User
Super User

Please see this post regarding How to Get Your Question Answered Quickly: https://community.powerbi.com/t5/Community-Blog/How-to-Get-Your-Question-Answered-Quickly/ba-p/38490


@ me in replies or I'll lose your thread!!!
Instead of a Kudo, please vote for this idea
Become an expert!: Enterprise DNA
External Tools: MSHGQM
YouTube Channel!: Microsoft Hates Greg
Latest book!:
Mastering Power BI 2nd Edition

DAX is easy, CALCULATE makes DAX hard...

Helpful resources

Announcements
April AMA free

Microsoft Fabric AMA Livestream

Join us Tuesday, April 09, 9:00 – 10:00 AM PST for a live, expert-led Q&A session on all things Microsoft Fabric!

March Fabric Community Update

Fabric Community Update - March 2024

Find out what's new and trending in the Fabric Community.