Hi there,
Just asking the community because I have just completed the lesson on Handling Missing Data in the Data Scientist:ML stream, and I’m a little unsure about the distinction between data missing at random vs data missing not at random. It seems a subtle difference but I can’t really understand why you would categorise data MAR given the examples provided and not MNAR?
Any insight would be appreciated,
Thanks:)
Perhaps this would help?
or,