Questions tagged [mnar]
MNAR is an acronym for "missing not at random" which, in the context of missing data, refers to data that is missing for reasons that are related to the missingness itself, in ways that cannot be explained by data that is present.
10
questions
2
votes
2
answers
35
views
Does a relationship between missing values indicate MNAR?
I have a dataset where several features are missing values only if another feature is also missing a value. Does this indicate that the missing data is missing not at random (MNAR)? Additionally, how ...
- 55
0
votes
0
answers
51
views
Handling a dependent null in linear regression
I have a linear regression model where I predict a user's completion rate for a specific game. The training data looks like this:
User Predictor variables
Game Predictor Variables
Completion rate of ...
2
votes
0
answers
14
views
MNAR Imputation On Time Series Data in Ongoing Study
I am working on an ongoing study that tracks subjects through time. Each subject is enrolled for months, where upon exiting I calculate their length-of-stay.
Since the study is ongoing, I have both ...
- 21
0
votes
0
answers
43
views
What are the implications for performing chi squared tests on ranking data?
I am working with a food preference dataset where individuals rank their food preferences (apple, banana, orange, coconut, pear). The goal is to determine whether one fruit is ranked as 'favorite' ...
- 305
4
votes
1
answer
86
views
Accuracy from biased sampling of boomerang throws
Suppose I keep throwing boomerangs every 10 seconds, and I throw 50,000 of them. 20% of the boomerangs are hit by drones so they don’t come back. The rest come back but the time it takes for them to ...
- 41
3
votes
2
answers
338
views
Techniques for MNAR without Imputation
Oftentimes I see that people address missingness from the perspective of missing at random (MAR). However, there are cases where missing not at random (MNAR) in which one can sometimes be handled with ...
- 8,655
-1
votes
1
answer
74
views
Question about MCAR, MAR and MNAR
I have 3 datasets. Each of the three datasets “data1.csv”, “data2.csv”, “data3.csv” is about the same sample of 654. (I imported the datasets in R studio.)
subjects: youths, aged 3 to 19
with the ...
0
votes
1
answer
288
views
How can I simulate NMAR missing data with bias?
Question summary
I am trying to simulate several random variables (using R), then introduce missingness into one of them, and show that a complete-case regression model yields biased results when the ...
1
vote
0
answers
37
views
Methods to account/correct for dropouts in intervention study
I have depression scores for patients before and after intervention, there is no control sample. I want to assess if demographics have an effect on recovery, and I was thinking of doing a linear ...
- 41
1
vote
1
answer
50
views
Can I correct for randomly missing data where missingness is has a known relationship to the error term?
Suppose I have a population of observations I want to model as being drawn from some distributional family, which I believe adequately represents the true distribution. My goal is to estimate the ...
- 3,047