Evolution Index On Matrix Visual

[vytautas1410]

Hello  Everyone, 

I have data of 2018 2019 2020 . I would like to see EVOLUTION INDEX of (2019 and 2020) on matrix visual, using (EINDEX = DIVIDE(1 + [Growth YOY],CALCULATE( 1 + [Growth YOY],ALLSELECTED())) * 100) measure. 

Currently I'm using this measure on Growth YOY:

Growth YOY =

VAR __PREV_YEAR = CALCULATE(SUM('M'[Value]), ALL('Calendar'[Date]) , DATEADD('Calendar'[Date].[Date], -1, YEAR))

VAR __THIS_YEAR = CALCULATE(SUM('M'[Value]),ALL('Calendar'[Date]) ,DATEADD('Calendar'[Date].[Date], 0, YEAR))

    RETURN

IF(__THIS_YEAR, (DIVIDE(SUM('M'[Value]) - __PREV_YEAR, __PREV_YEAR))) 

I'm only getting Evolution INDEX on last 12 months , and when I'm filtering by Date, the values are slightly changing.

I believe that the problem is with GROWTH YOY measure .  Could someone help me with these measures ? 

Thank you in advance . 

[CNENFRNL]

@vytautas1410 Indeed, it requires a glance of data model to troubleshoot. Btw, I think in such expressions, ALL('Calendar'[Date]) is redundant.

VAR __PREV_YEAR = CALCULATE(SUM('M'[Value]), ALL('Calendar'[Date]) , DATEADD('Calendar'[Date].[Date], -1, YEAR))
VAR __THIS_YEAR = CALCULATE(SUM('M'[Value]),ALL('Calendar'[Date]) ,DATEADD('Calendar'[Date].[Date], 0, YEAR))

 When the first parameter of DATEADD func references a date column, it's replaced by CALCULATETABLE ( DISTINCT ( <Dates> ) ); context transition applies and filter context also takes effect accordingly. DATEADD returns a date table after the call, which is surely a subset of ALL(Calendar[Date]).

[Greg_Deckler]

@vytautas1410 Sorry, having trouble following, can you post sample data as text and expected output?

Not really enough information to go on, please first check if your issue is a common issue listed here: https://community.powerbi.com/t5/Community-Blog/Before-You-Post-Read-This/ba-p/1116882

Also, 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

The most important parts are:
1. Sample data as text, use the table tool in the editing bar
2. Expected output from sample data
3. Explanation in words of how to get from 1. to 2.