Questions tagged [missing-data]
When the data present lack of information (gaps), i.e., are not complete. Hence, it is important to consider this feature when performing an analysis or test.
1,686 questions
0
votes
0
answers
18
views
Meta-regression with missing values
I want to perform a meta-regression to explore the sources of heterogeneity in my meta-analysis and estimate the impact of covariates on effect size.
However, if the value of a covariate is missing ...
- 83
2
votes
2
answers
70
views
Lab results missing due to technical failures: can this be treated as MCAR?
In lab data, most missingness seems due to technical/operational failures (no draw, sample error, insufficient volume, lost/mislabeled tube or reading error due to label printing), so I’m inclined to ...
- 869
6
votes
1
answer
189
views
Handling missing values of categorical variables in medical datasets
I am working with medical datasets that represent open-source time-to-event datasets. These are numerical datasets and one one sample dataset is presented here:
As, you can see that, in the dataset, ...
8
votes
2
answers
133
views
Multiple Imputation from only demographics
I am using multiple imputation to handle missing values in a data set. The question is what to do about people who answered the demographic questions, but then dropped out. So all I have are age, race,...
- 386
0
votes
0
answers
12
views
Multiple imputations for missing data in repeated measures design where some questions are not asked across time points
I have data from surveys pre-, during (just for some feedback on the intervention itself), and post-intervention. I have stacked the data for each time point in long format. Questions in the during-...
1
vote
1
answer
74
views
Is it appropriate to keep an interaction term in a Type II ANOVA if it is non-significant in a Type I ANOVA?
I fitted a Type I ANOVA (Type I SS) model with two main effects and their interaction, but then realized that one group has a missing data point. In the Type I ANOVA, the interaction term was not ...
- 385
0
votes
0
answers
28
views
R - Interrater-Reliability (each object rated by different raters)
I am working on the psychometric evaluation of a questionnaire. In this questionnaire employees are rating the sites they work at. So I am interested in interrater-reliability, that is how much do ...
- 1
0
votes
0
answers
44
views
How to impute the data under specific constraints?
Medicine A (Received or not)
Dose (unit, RANGE:[10-90])
Y
15
N
Not on Medicine A
Y
15
Y
60
Y
90
Y
18
Y
-99
N
Not on Medicine A
So here, Medicine A is the indicator to indicate whether the patients ...
- 528
2
votes
0
answers
42
views
Restrict training data to only rows with values for most important variable? [closed]
My training data is mostly missing values for the feature that I know will be the most important variable. This missingness is semi-random. For example, I know the value is missing for this feature ...
- 237
2
votes
0
answers
49
views
How to handle SEER “No/Unknown” treatment coding in Cox models (train on SEER, validate on external data)?
Context: I’m building an overall-survival Cox model for CNS tumors trained on SEER and validated on my institution’s registry. SEER records chemotherapy and radiotherapy as Yes vs No/Unknown (cannot ...
- 373
3
votes
1
answer
149
views
Bayesian Clustering with a Finite Gaussian Mixture Model with Missing Data
I would like to perform clustering with a finite Gaussian Mixture model, however, I have missing data (some features are missing at random). I am using Variational Inference to fit my Bayesian GMM. Is ...
- 1,112
4
votes
1
answer
97
views
Imputation of Data for CFA
I have a dataset with some latent variables, and my main one happens to have 9 dichotomous items.
I did little MCAR's test which resulted in a very low p-value, so I should conduct imputation before ...
- 43
4
votes
2
answers
301
views
Use bigger sample for predictors in regression
I need to regress continuous y on multi-dimensional X (for prediction mostly, not inference, but do I need the betas to make ...
- 141
5
votes
1
answer
171
views
How to handle missing weather data (predictors) in production time series forecast with Prophet?
I'm currently running a production pipeline that uses Facebook Prophet (GAM) to forecast future electricity usage. The model includes:
Target: past electricity consumption (hourly data)
External ...
- 53
1
vote
1
answer
94
views
Poisson Mixed_effects and Maximum-Likelihood estimation with missing data
I'm running an analysis using both linear mixed effects models (for my continuous outcomes) and generalized linear mixed effects models (for my binary outcomes). I have two questions:
When using a ...
- 11