A comparison of missing data approaches for linear regression with missing not at random outcome and predictors
{{output}}
While most methods for missing not at random (MNAR) data in regression models address MNAR outcomes assuming fully observed predictors, real-world observational health and longitudinal studies often violate this assumption. This paper compares several approach... ...
