Forum Discussion
Detect changed value in a column compared to the previous entry for the same id.
Hey Team,
I have a dataset that has numerous features, but for indexing purposes
| globalid | editdate | flag_latest | counter | previous editdate | rest of features |
globalid - user_id, can appear numerous times.
editdate - date of modification to any of the features
previous editdate - lag (editdate)
flag_latest - latest editdate = 1, everything else = 0
counter - number of times the globalid has appeared
Data changes through operational use, but we are also doing a lot of "cleaning" of the data, and I want to be able to measure change in features in order to evaluate what it is exactly that we are cleaning.
For example:
I know feature 1 is a status, so I'm ok with it changing based on different operational events.
But feature 5 is an initial value and it should be alarming if that changes.
So I would be able to create a visualization of this sort:
For that, I understand that what I'm after is the "change in a feature".
For this, I'd need an if statement that identifies a change in the feature against the previous value of the same globalid.
My question is if I have a way with DAX measures to not have to go through an if statement for every single column (feature).
1 Reply
- Greg_Deckler
Community Champion
asp_nizcovich Seems like MTBF. I can't really understand what your source data looks like.
See my article on Mean Time Between Failure (MTBF) which uses EARLIER: http://community.powerbi.com/t5/Community-Blog/Mean-Time-Between-Failure-MTBF-and-Power-BI/ba-p/339586.
The basic pattern is:
Column =
VAR __Current = [Value]
VAR __PreviousDate = MAXX(FILTER('Table','Table'[Date] < EARLIER('Table'[Date])),[Date])
VAR __Previous = MAXX(FILTER('Table',[Date]=__PreviousDate),[Value])
RETURN
__Current - __PreviousBut if that is not helpful, 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.
